初始化项目，由ModelHub XC社区提供模型

Model: pbhappliedsystems/qwen-2.5-14B-instruct-1m-gguf-F16 Source: Original Platform
2026-07-25 09:11:10 +08:00
commit cd073b6362
7 changed files with 1528 additions and 0 deletions
--- a/.gitattributes
+++ b/.gitattributes
@@ -0,0 +1,36 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
+qwen-2.5-14B-instruct-1m-gguf-F16.gguf filter=lfs diff=lfs merge=lfs -text
--- a/LICENSE_Apache-2.0
+++ b/LICENSE_Apache-2.0
@@ -0,0 +1,170 @@
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship made available under
+      the License, as indicated by a copyright notice that is included in
+      or attached to the work (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean, as submitted to the Licensor for inclusion
+      in the Work by the copyright owner or by an individual or Legal Entity
+      authorized to submit on behalf of the copyright owner. For the purposes
+      of this definition, "submitted" means any form of electronic, verbal, or
+      written communication sent to the Licensor or its representatives,
+      including but not limited to communication on electronic mailing lists,
+      source code control systems, and issue tracking systems that are managed
+      by, or on behalf of, the Licensor for the purpose of discussing and
+      improving the Work, but excluding communication that is conspicuously
+      marked or otherwise designated in writing by the copyright owner as
+      "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any Legal Entity on behalf of
+      whom a Contribution has been received by the Licensor and subsequently
+      incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a cross-claim
+      or counterclaim in a lawsuit) alleging that the Work or a Contribution
+      incorporated within the Work constitutes direct or contributory patent
+      infringement, then any patent licenses granted to You under this License
+      for that Work shall terminate as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or Derivative Works
+          a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, You must include a readable copy of the attribution
+          notices contained within such NOTICE file, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own license statement for Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such derivative works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or exemplary damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or all other
+      commercial damages or losses), even if such Contributor has been
+      advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
--- a/README.md
+++ b/README.md
@@ -0,0 +1,390 @@
+---
+language:
+  - en
+  - zh
+  - fr
+  - es
+  - pt
+  - de
+  - it
+  - ru
+  - ja
+  - ko
+  - ar
+license: apache-2.0
+base_model: Qwen/Qwen2.5-14B-Instruct-1M
+tags:
+  - gguf
+  - f16
+  - full-precision
+  - qwen
+  - qwen2.5
+  - instruct
+  - llama-cpp
+  - long-context
+  - 1m-context
+  - agentic
+  - structured-output
+  - pbh-applied-systems
+  - quant-eval
+  - baseline
+---
+
+# Qwen2.5-14B-Instruct-1M · GGUF F16
+
+**Converted and evaluated by [PBH Applied Systems, LLC](https://pbhappliedsystems.com)**
+— Applied AI/ML Consulting · LLM Optimization & Deployment · Quantized AI Infrastructure
+
+> 🔬 **This repository is part of a production-oriented evaluation series.** Every model published under [`pbhappliedsystems`](https://huggingface.co/pbhappliedsystems) has been independently evaluated using **quant_eval v7.21** — a proprietary behavioral evaluation harness developed by PBH Applied Systems.
+
+> 📌 **This is the full-precision F16 baseline repository.** The evaluated Q4\_K\_M deployment variant is published at [`pbhappliedsystems/qwen-2.5-14B-instruct-1m-gguf-Q4-K-M`](https://huggingface.co/pbhappliedsystems/qwen-2.5-14B-instruct-1m-gguf-Q4-K-M). That card documents the complete cross-series comparisons, context window VRAM guide, and deployment recommendations. **The Q4\_K\_M variant is the recommended choice for all deployments** — it achieves identical behavioral results at 21.1× faster inference.
+
+---
+
+## Try the Live AI Agent Demo
+
+[**Launch the PBH Applied Systems AI Agent Demo →**](https://pbhappliedsystems.com/assistant.html)
+
+This model is part of the PBH Applied Systems evaluated model series that supports the live AI Agent Demo. The demo lets visitors interact with production-style agent workflows powered by open-weight language models evaluated through PBH Applied Systems' `quant_eval` framework.
+
+The F16 model serves a different role than the Q4_K_M deployment variant. F16 is the full-precision baseline used to measure what the model can do before quantization. `quant_eval` then compares the quantized model against this baseline to identify which capabilities are preserved, which degrade, and which tasks require guardrails or a higher-precision deployment.
+
+This comparison is central to the demo. It helps determine which model belongs in which agent role:
+
+- Reasoning models are selected for planning, analysis, and auditable decision workflows.
+- Document models are selected for long-context extraction, summarization, and structured Q&A.
+- Code models are selected for task completion, structured output, API scaffolding, and automation workflows.
+- Quantized variants are selected when they preserve enough behavior to reduce cost, latency, and GPU requirements.
+- F16 variants remain important when maximum fidelity, cleaner tool execution, or reduced quantization risk matters more than speed or cost.
+
+The live demo shows the deployment side of that process. The F16 card documents the reference behavior. The Q4_K_M card shows what changes after compression. Together, they explain how PBH Applied Systems uses `quant_eval` to choose the correct LLM for the correct agent type instead of guessing from model size or leaderboard reputation.
+
+---
+
+## Model Description
+
+This repository contains the **full-precision F16 GGUF** of [`Qwen/Qwen2.5-14B-Instruct-1M`](https://huggingface.co/Qwen/Qwen2.5-14B-Instruct-1M), a 14-billion parameter instruction-tuned model from Alibaba Cloud featuring a **1,000,000-token context window**.
+
+In the PBH Applied Systems evaluation pipeline, this F16 run (`20260210_215029`) operated in cache-generation mode (`skip_quant=true`), producing the `full_weight_cache.json` used as the reference baseline for the subsequent Q4\_K\_M comparison run (`20260210_235131`). The evaluation results here are the source of the F16 baseline data shown in the Q4\_K\_M card — timing profiles and raw outputs are identical across both runs, confirming clean cache reuse and full run integrity.
+
+### Key Characteristics
+
+- **Parameters:** 14B
+- **Format:** GGUF F16 (full precision)
+- **File size:** 29.5 GB
+- **SHA256:** `de08ea9c41234ef83b7aacf07f9ebc3cbaa20ca8aeb5f6417758a8798660aaa9`
+- **Context window:** **1,000,000 tokens**
+- **Minimum VRAM (GPU inference):** ~32 GB (short context) — scales with context length
+- **Recommended GPU tier:** A100 40 GB · 2× RTX 4090
+- **Inference speed (eval hardware):** avg **56.623 sec/case** on RTX 4090
+- **License:** Apache 2.0
+
+> **On inference speed:** The F16 model averages 56.6 sec/case — nearly one minute per structured inference task on an RTX 4090. The json\_01 case takes **376.99 seconds** (over 6 minutes). For this model, the **Q4\_K\_M variant (2.683 sec/case average)** is the operationally viable choice on all but the highest-VRAM multi-GPU setups. Both produce identical behavioral results.
+
+---
+
+## PBH Applied Systems Evaluation — quant\_eval v7.21
+
+> **Evaluation conducted by PBH Applied Systems, LLC using quant_eval v7.21**
+> Run ID: `20260210_215029` · Fixtures: `golden_oracle_fixtures_v7_21` (SHA256: `6d71a0b9147c...`) · Seed: 42
+> Hardware: NVIDIA RTX 4090 · Runner: `full_weight_transformers` (F16 only) · Total rows: 42
+
+### Per-Family Pass Rates — F16 (`full_weight_transformers`)
+
+| Family | N | Pass Rate | Avg Secs | Bucket Score | Q4\_K\_M Parity |
+|---|---:|---:|---:|---:|---|
+| json\_multistep | 5 | **0.800** | 133.21 | 2.200 | ✅ Identical |
+| stateful\_followup | 2 | **1.000** | 13.75 | 2.000 | ✅ Identical |
+| toolcall\_only | 2 | 0.000 | 15.36 | 1.000 | ✅ Identical |
+| mixed\_brief\_json | 2 | **1.000** | 19.40 | 2.000 | ✅ Identical |
+| toolcall | 2 | **1.000** | 25.95 | **11.000** | ✅ Identical |
+| json | 4 | n/a | 143.87 | 10.000 | ✅ Identical |
+| fuzz | 20 | n/a | 49.21 | 10.000 | ✅ Identical |
+| mcq | 5 | n/a | 0.73 | **1.000** | ✅ Identical |
+
+**Every family result is identical between F16 and Q4\_K\_M.** This model is the only one in the evaluated series with zero measurable quantization degradation across all behavioral families.
+
+---
+
+## F16-Specific Observations
+
+### toolcall — bucket=11, Clean Final Answers at F16
+
+Both `toolcall` cases pass with the maximum bucket score at F16 — no role-token contamination, no EOS tokens, no missing answers:
+
+| Case | Raw Output | Expected | Result |
+|---|---|---|---|
+| tool\_01 | `{...add(2,3)...}  5` | `5` | ✅ bucket=11 |
+| tool\_02 | `{...add(10,-4)...}  6` | `6` | ✅ bucket=11 |
+
+This matches the Q4\_K\_M runner exactly. Qwen2.5-14B-Instruct-1M at F16 does not exhibit the role-token contamination (PARTICULAR: annotation, garbled prefixes) documented in the Qwen2.5-7B F16 evaluation, nor the EOS contamination of the smaller Qwen Q4\_K\_M variants. Clean output requires no post-processing at this precision level.
+
+### toolcall\_only — "left"/"right" Schema Vocabulary at F16
+
+Both `toolcall_only` cases use `"left"/"right"` as argument keys — the same vocabulary as the F16 runner for this model:
+
+| Case | Raw Output |
+|---|---|
+| toolonly\_01 | `{"tool": "add", "left": 5, "right": 10}` |
+| toolonly\_02 | `{"tool": "add", "left": 25, "right": 75}` |
+
+Contrast with Q4\_K\_M which uses a nested `"input"` object. Both runners fail `args_ok` but with different wrong schemas. Explicit key names in the system prompt resolve this at both precision levels.
+
+### MCQ — Perfect 5/5 at F16
+
+All five MCQ cases return a clean single-character answer in ~0.73 seconds:
+
+```
+mcq_01: B  |  mcq_02: B  |  mcq_03: C  |  mcq_04: B  |  mcq_05: B
+```
+
+No empty output, no invalid choices, no A-bias. At F16 MCQ is the fastest family in this run — the 1M context window doesn't affect short-response tasks.
+
+### json\_01 — 376.99-Second Outlier
+
+`json_01` at F16 takes **376.99 seconds** while json\_02 through json\_04 run 63–70 seconds each. The output is correct (bucket=10). This extreme variance is a characteristic of the 1M context window at full precision — certain inputs trigger substantially longer generation sequences before the model settles on its brief JSON output. The Q4\_K\_M runner takes 3.38 seconds on the same case, confirming this is a precision + context-window interaction, not a fixture complexity issue.
+
+### stateful\_followup — No Turn-2 Contamination
+
+Unlike the Qwen2.5-7B F16 evaluation (which showed `PARTICULAR:` annotation hallucinations on turn-2), this model produces clean JSON state at both turns with no appended text:
+
+| Case | Turn 1 | Turn 2 |
+|---|---|---|
+| state\_01 | `{"counter": 2}` | `{"counter": 5}` |
+| state\_02 | `{"items": ["a", "b"]}` | `{"items": ["a", "b", "c"]}` |
+
+The F16 Transformers runner handles this model's stateful outputs cleanly.
+
+---
+
+## F16 vs. Q4\_K\_M — Deployment Decision
+
+| Dimension | F16 (this repo) | Q4\_K\_M |
+|---|---|---|
+| VRAM (8K context) | ~32 GB | ~12 GB |
+| VRAM (128K context) | ~48 GB | ~20 GB |
+| Avg inference time | 56.623 sec/case | 2.683 sec/case |
+| Speed ratio | 1.0× (baseline) | **21.1× faster** |
+| All family pass rates | Same | Same |
+| Toolcall final answer | Clean (bucket=11) | Clean (bucket=11) |
+| MCQ | 5/5 | 5/5 |
+| Behavioral difference | None | None |
+
+**For every practical deployment scenario, Q4\_K\_M is the correct choice.** It achieves the same results in 21× less time at ~3× less VRAM. F16 is appropriate only when: (1) you have 32+ GB GPU VRAM available, (2) you require full-weight provenance for compliance or reproducibility auditing, or (3) you need the F16 baseline cache for a subsequent comparison evaluation run.
+
+---
+
+## Hardware Requirements
+
+| Configuration | VRAM Required | Notes |
+|---|---|---|
+| F16 (this repo) · 8K context | ~32 GB | 29.5 GB model + KV cache |
+| F16 · 32K context | ~36 GB | Minimum A100 40 GB |
+| F16 · 128K context | ~48 GB | A100 80 GB or multi-GPU |
+| Q4\_K\_M (companion repo) · 8K | ~12 GB | 8.99 GB model + KV cache |
+| Q4\_K\_M · 128K context | ~20 GB | A10G 24 GB · RTX 4090 |
+
+---
+
+## Usage
+
+### Installation
+
+```bash
+pip install llama-cpp-python huggingface_hub
+```
+
+For GPU acceleration (CUDA):
+
+```bash
+CMAKE_ARGS="-DGGML_CUDA=on" pip install llama-cpp-python --force-reinstall --no-cache-dir
+```
+
+### Python — llama-cpp-python
+
+```python
+from huggingface_hub import hf_hub_download
+from llama_cpp import Llama
+
+# Note: 29.5 GB download — ensure sufficient disk space and ~32 GB VRAM
+model_path = hf_hub_download(
+    repo_id="pbhappliedsystems/qwen-2.5-14B-instruct-1m-gguf-F16",
+    filename="qwen-2.5-14B-instruct-1m-gguf-F16.gguf"
+)
+
+llm = Llama(
+    model_path=model_path,
+    n_ctx=32768,      # Set to actual working context; supports up to 1M
+    n_gpu_layers=-1,
+    verbose=False,
+)
+
+response = llm.create_chat_completion(
+    messages=[
+        {
+            "role": "system",
+            "content": "You are a precise assistant. Follow instructions exactly."
+        },
+        {
+            "role": "user",
+            "content": "Analyze the following and return a JSON object with keys: summary, risk_level, action_items."
+        }
+    ],
+    temperature=0.7,
+    max_tokens=1024,
+)
+print(response["choices"][0]["message"]["content"])
+```
+
+For tool-calling (no EOS stripping required — output is clean at F16):
+
+```python
+# quant_eval v7.21: toolcall bucket=11 — clean final answer, no post-processing needed
+response = llm.create_chat_completion(
+    messages=[
+        {
+            "role": "system",
+            "content": (
+                "You are a tool-calling assistant. Output the tool call as JSON, "
+                "then on the next line output only the numeric result.\n"
+                'Tool call format: {"tool_name": "<n>", "args": {"a": <n>, "b": <n>}}'
+            )
+        },
+        {"role": "user", "content": "Use the add tool to compute 10 minus 4."}
+    ],
+    temperature=0.7,
+    max_tokens=128,
+)
+# No stripping required at F16 — output is clean
+print(response["choices"][0]["message"]["content"])
+```
+
+For bare tool-call dispatch with schema enforcement:
+
+```python
+import json, re
+
+def call_tool_bare(llm, prompt: str, retries: int = 3) -> dict:
+    """
+    Explicit schema enforcement for toolcall_only.
+    quant_eval v7.21: F16 uses 'left'/'right' keys without schema guidance.
+    System prompt specifying exact keys resolves the vocabulary mismatch.
+    """
+    for attempt in range(retries):
+        response = llm.create_chat_completion(
+            messages=[
+                {
+                    "role": "system",
+                    "content": (
+                        'Respond ONLY with a JSON object using EXACTLY these keys:\n'
+                        '{"tool_name": "add", "args": {"a": <integer>, "b": <integer>}}\n'
+                        'No other text, no markdown.'
+                    )
+                },
+                {"role": "user", "content": prompt}
+            ],
+            temperature=0.0,
+            max_tokens=64,
+        )
+        raw = response["choices"][0]["message"]["content"].strip()
+        try:
+            parsed = json.loads(raw)
+            assert "tool_name" in parsed and "args" in parsed
+            assert "a" in parsed["args"] and "b" in parsed["args"]
+            return parsed
+        except (json.JSONDecodeError, AssertionError, KeyError):
+            if attempt == retries - 1:
+                raise ValueError(f"Tool call failed after {retries} attempts. Raw: {raw}")
+```
+
+### CLI — llama-cli
+
+```bash
+llama-cli \
+  --model qwen-2.5-14B-instruct-1m-gguf-F16.gguf \
+  --chat-template qwen2 \
+  --system-prompt "You are a precise assistant. Follow instructions exactly." \
+  --prompt "Return a JSON object with keys: summary, risk_level, action_items." \
+  --n-predict 1024 \
+  --ctx-size 32768 \
+  --n-gpu-layers -1 \
+  --temp 0.7
+```
+
+---
+
+## Artifact Provenance
+
+| Artifact | Format | Size | SHA256 |
+|---|---|---|---|
+| `qwen-2.5-14B-instruct-1m-gguf-F16.gguf` | GGUF F16 | 29.5 GB | `de08ea9c41234ef83b7aacf07f9ebc3cbaa20ca8aeb5f6417758a8798660aaa9` |
+| Q4\_K\_M *(companion repo)* | GGUF Q4\_K\_M | 8.99 GB | `5ad529ff2b1b192f31c8a638fe8756a0c628904e2ded797c11f9194216976973` |
+
+The F16 GGUF was converted from `Qwen/Qwen2.5-14B-Instruct-1M` using a custom-built llama.cpp conversion pipeline developed by PBH Applied Systems.
+
+**Two-pass architecture:** This F16 run (`20260210_215029`) operated in cache-generation mode (`skip_quant=true`). The resulting `full_weight_cache.json` was used as the reference baseline for the Q4\_K\_M comparison run (`20260210_235131`). Timing identity between this run and the F16 baseline entries in the comparison run confirms clean cache reuse and run integrity.
+
+---
+
+## Evaluation Methodology
+
+**quant_eval v7.21** — proprietary behavioral evaluation harness, PBH Applied Systems.
+**Fixture set:** `golden_oracle_fixtures_v7_21` (SHA256: `6d71a0b9147c079371b02a94f3c149eb78a6adc03dc16ff6833b964fbf4174f0`)
+**Evaluation hardware:** NVIDIA RTX 4090 · **F16 evaluation date:** February 10, 2026 · **Seed:** 42
+
+---
+
+## 🔬 About quant_eval & This Evaluation Series
+
+[**quant_eval**](https://pbhappliedsystems.com) is a proprietary behavioral evaluation harness developed by [PBH Applied Systems, LLC](https://pbhappliedsystems.com). It measures real agent-adjacent task performance across structured output, tool dispatch, multi-turn state retention, and multi-step planning — not perplexity or leaderboard proxies. Every model published under [`pbhappliedsystems`](https://huggingface.co/pbhappliedsystems) has been independently evaluated using quant_eval before being recommended for any production role.
+
+**See it in action:** [**Live AI Agent Demo →**](https://pbhappliedsystems.com/assistant.html)
+The demo runs production-style agent workflows powered by open-weight models selected through the quant_eval evaluation pipeline.
+
+> **Need a deployment recommendation?**
+> Not sure which quantization level is right for your hardware, latency target, or agent type?
+> [**→ pbhappliedsystems.com**](https://pbhappliedsystems.com)
+
+---
+
+*Evaluated and published by [PBH Applied Systems, LLC](https://pbhappliedsystems.com) · [patrick@pbhappliedsystems.com](mailto:patrick@pbhappliedsystems.com)*
+
+---
+
+## About PBH Applied Systems
+
+[**PBH Applied Systems, LLC**](https://pbhappliedsystems.com) is an Oklahoma City–based applied machine learning and AI systems company specializing in production-grade model evaluation, quantization pipelines, agentic AI infrastructure, and scalable AI-driven application development.
+
+**Patrick Hill, M.S.** — Founder · Data Scientist · AI/ML Engineer · Author of **[Applied Machine Learning: Concepts, Tools, and Case Studies](https://a.co/d/05qat7Xz)** (required reading, UAT CSC 373)
+
+---
+
+## 📞 Work With PBH Applied Systems
+
+The F16 baseline for this model exists because proper quantization evaluation requires it — you cannot measure what Q4\_K\_M preserves or degrades without a verified full-precision reference. For this model, the answer is: zero degradation. That finding only becomes a deployable fact when both runs exist and can be compared against the same fixture set.
+
+👉 **[Book a Scoping Call](https://pbhappliedsystems.com)** · 👉 **[Request an Evaluation Report](https://pbhappliedsystems.com)** — from $2,500
+
+### Connect
+
+| | |
+|---|---|
+| 🌐 | [pbhappliedsystems.com](https://pbhappliedsystems.com) |
+| 📧 | [patrick@pbhappliedsystems.com](mailto:patrick@pbhappliedsystems.com) |
+| 💼 | [LinkedIn](https://www.linkedin.com/company/pbh-applied-systems-llc) |
+| ▶️ | [YouTube](https://www.youtube.com/@pbhappliedsystems) |
+| 📸 | [Instagram](https://www.instagram.com/pbhappliedsystems) |
+| 👍 | [Facebook](https://www.facebook.com/pbhappliedsystems) |
+
+---
+
+## License
+
+This GGUF repository inherits the license of the base model:
+**Apache 2.0** — [`Qwen/Qwen2.5-14B-Instruct-1M`](https://huggingface.co/Qwen/Qwen2.5-14B-Instruct-1M)
+
+The quant_eval evaluation methodology, fixture set, and scoring framework are proprietary to PBH Applied Systems, LLC and are not included in this repository.
+
+---
+
+*GGUF conversion and behavioral evaluation performed by [PBH Applied Systems, LLC](https://pbhappliedsystems.com) · quant_eval v7.21 · F16 Run ID: `20260210_215029`*
--- a/comparison_results_v7_21_Qwen2.5_14B_Instruct_1M_20260210_215029.csv
+++ b/comparison_results_v7_21_Qwen2.5_14B_Instruct_1M_20260210_215029.csv
@@ -0,0 +1,630 @@
+run_id,timestamp,runner,family,case_id,mode,secs,timeout,output_raw,error,plan_len,plan_exact_match,plan_prefix_match_len,plan_stop_idx_expected,plan_stop_idx_got,final_equiv_ok,final_constraints_ok,bucket_score,detail,expected_plan,got_plan,expected_final,got_final,answer_line_ok,args_ok,best_of_k_candidate_idx,best_of_k_k,best_of_k_runner,best_of_k_seed,best_of_k_selected_idx,best_of_k_selected_sim,best_of_k_shot1_tier1_pass,best_of_k_tier1_pass_at_k,best_of_k_used,checks_consistent_ok,checks_ok,choice_extracted,constraints_detail,detail_stage1,expected_answer_line,expected_args,expected_json,expected_turn1,expected_turn2,final_consistent_ok,final_match_reported,got_answer_line,got_args,got_json,got_turn1,got_turn2,json_exact_match,json_parse_ok,oracle_equiv_ok,oracle_trace,parse_ok,plan_len_ok,schema_errors,schema_ok,stage1_tool_parse_ok,stage1_tool_schema_ok,stop_semantics_ok,tool_exec_detail,tool_exec_ok,tool_name_ok,tool_output_json,tool_payload_used,turn1_exact_match,turn1_parse_ok,turn2_exact_match,turn2_parse_ok
+20260210_215029,20260210_220525,full_weight_transformers,json,json_01,tool_decisions,376.99,0,"{
+  ""tool_name"": ""place_item"",
+  ""args"": {
+    ""choice"": ""A""
+  }
+}
+
+{
+  ""tool_name"": ""place_item"",
+  ""args"": {
+    ""choice"": ""A""
+  }
+}
+
+{
+  ""tool_name"": ""place_item"",
+  ""args"": {
+    ""choice"": ""C""
+  }
+}",,3,1,3,-1,-1,1,1,10,ok,"[""A"",""A"",""C""]","[""A"",""A"",""C""]","{""A"":[3,3,0],""B"":[2,0,0],""C"":[0,0,2],""new_shelf"":0,""violation"":1}","{""A"":[3,3,0],""B"":[2,0,0],""C"":[0,0,2],""new_shelf"":0,""violation"":1}",,,,,,,,,,,,,,,ok,,,,,,,,,,,,,,,,,"Initial A=[2, 2, 0] B=[2, 0, 0] C=[0, 0, 1]
+Arrivals=['P', 'S', 'H']
+Arrival 1: P
+  Try A: can_place=1 state=[2, 2, 0]
+  Placed on A: new_state=[2, 3, 0]
+Arrival 2: S
+  Try A: can_place=1 state=[2, 3, 0]
+  Placed on A: new_state=[3, 3, 0]
+Arrival 3: H
+  Try A: can_place=0 state=[3, 3, 0]
+  Try B: can_place=0 state=[2, 0, 0]
+  Try C: can_place=1 state=[0, 0, 1]
+  Placed on C: new_state=[0, 0, 2]
+Final A=[3, 3, 0] B=[2, 0, 0] C=[0, 0, 2] new_shelf=0 violation=1",,,,,,,,,,,,,,,,
+20260210_215029,20260210_220525,full_weight_transformers,json,json_02,tool_decisions,63.22,0,"{
+  ""tool_name"": ""place_item"",
+  ""args"": {
+    ""choice"": ""A""
+  }
+}
+
+{
+  ""tool_name"": ""place_item"",
+  ""args"": {
+    ""choice"": ""A""
+  }
+}
+
+{
+  ""tool_name"": ""place_item"",
+  ""args"": {
+    ""choice"": ""A""
+  }
+}",,3,1,3,-1,-1,1,1,10,ok,"[""A"",""A"",""A""]","[""A"",""A"",""A""]","{""A"":[3,1,0],""B"":[0,1,0],""C"":[0,0,1],""new_shelf"":0,""violation"":1}","{""A"":[3,1,0],""B"":[0,1,0],""C"":[0,0,1],""new_shelf"":0,""violation"":1}",,,,,,,,,,,,,,,ok,,,,,,,,,,,,,,,,,"Initial A=[1, 0, 0] B=[0, 1, 0] C=[0, 0, 1]
+Arrivals=['S', 'S', 'P']
+Arrival 1: S
+  Try A: can_place=1 state=[1, 0, 0]
+  Placed on A: new_state=[2, 0, 0]
+Arrival 2: S
+  Try A: can_place=1 state=[2, 0, 0]
+  Placed on A: new_state=[3, 0, 0]
+Arrival 3: P
+  Try A: can_place=1 state=[3, 0, 0]
+  Placed on A: new_state=[3, 1, 0]
+Final A=[3, 1, 0] B=[0, 1, 0] C=[0, 0, 1] new_shelf=0 violation=1",,,,,,,,,,,,,,,,
+20260210_215029,20260210_220525,full_weight_transformers,json,json_03,tool_decisions,64.75,0,"{
+  ""tool_name"": ""place_item"",
+  ""args"": {
+    ""choice"": ""C""
+  }
+}
+
+{
+  ""tool_name"": ""place_item"",
+  ""args"": {
+    ""choice"": ""C""
+  }
+}
+
+{
+  ""tool_name"": ""place_item"",
+  ""args"": {
+    ""choice"": ""A""
+  }
+}",,3,1,3,-1,-1,1,1,10,ok,"[""C"",""C"",""A""]","[""C"",""C"",""A""]","{""A"":[3,1,0],""B"":[0,3,0],""C"":[0,0,4],""new_shelf"":0,""violation"":1}","{""A"":[3,1,0],""B"":[0,3,0],""C"":[0,0,4],""new_shelf"":0,""violation"":1}",,,,,,,,,,,,,,,ok,,,,,,,,,,,,,,,,,"Initial A=[3, 0, 0] B=[0, 3, 0] C=[0, 0, 2]
+Arrivals=['H', 'H', 'P']
+Arrival 1: H
+  Try A: can_place=0 state=[3, 0, 0]
+  Try B: can_place=0 state=[0, 3, 0]
+  Try C: can_place=1 state=[0, 0, 2]
+  Placed on C: new_state=[0, 0, 3]
+Arrival 2: H
+  Try A: can_place=0 state=[3, 0, 0]
+  Try B: can_place=0 state=[0, 3, 0]
+  Try C: can_place=1 state=[0, 0, 3]
+  Placed on C: new_state=[0, 0, 4]
+Arrival 3: P
+  Try A: can_place=1 state=[3, 0, 0]
+  Placed on A: new_state=[3, 1, 0]
+Final A=[3, 1, 0] B=[0, 3, 0] C=[0, 0, 4] new_shelf=0 violation=1",,,,,,,,,,,,,,,,
+20260210_215029,20260210_220525,full_weight_transformers,json,json_04,tool_decisions,70.51,0,"{
+  ""tool_name"": ""place_item"",
+  ""args"": {
+    ""choice"": ""A""
+  }
+}
+
+{
+  ""tool_name"": ""place_item"",
+  ""args"": {
+    ""choice"": ""A""
+  }
+}
+
+{
+  ""tool_name"": ""place_item"",
+  ""args"": {
+    ""choice"": ""B""
+  }
+}",,3,1,3,-1,-1,1,1,10,ok,"[""A"",""A"",""B""]","[""A"",""A"",""B""]","{""A"":[1,1,0],""B"":[0,0,1],""C"":[0,0,0],""new_shelf"":0,""violation"":1}","{""A"":[1,1,0],""B"":[0,0,1],""C"":[0,0,0],""new_shelf"":0,""violation"":1}",,,,,,,,,,,,,,,ok,,,,,,,,,,,,,,,,,"Initial A=[0, 0, 0] B=[0, 0, 0] C=[0, 0, 0]
+Arrivals=['S', 'P', 'H']
+Arrival 1: S
+  Try A: can_place=1 state=[0, 0, 0]
+  Placed on A: new_state=[1, 0, 0]
+Arrival 2: P
+  Try A: can_place=1 state=[1, 0, 0]
+  Placed on A: new_state=[1, 1, 0]
+Arrival 3: H
+  Try A: can_place=0 state=[1, 1, 0]
+  Try B: can_place=1 state=[0, 0, 0]
+  Placed on B: new_state=[0, 0, 1]
+Final A=[1, 1, 0] B=[0, 0, 1] C=[0, 0, 0] new_shelf=0 violation=1",,,,,,,,,,,,,,,,
+20260210_215029,20260210_223824,full_weight_transformers,json_multistep,ms_easy_01,best_of_k,108.92995429039001,0,"{
+  ""plan"": [""A""],
+  ""checks"": [
+    {
+      ""step"": 1,
+      ""item"": ""P"",
+      ""choice"": ""A"",
+      ""can_place"": 1
+    }
+  ],
+  ""final"": {
+    ""A"": [2, 2, 1],
+    ""B"": [3, 0, 0],
+    ""C"": [0, 0, 1],
+    ""new_shelf"": 0,
+    ""violation"": 0
+  },
+  ""stop"": false,
+  ""stop_reason"": """",
+  ""stop_step_index"": -1
+}",,1,1,1,-1,-1,1,1,1,ok,"[""A""]","[""A""]","{""A"":[2,3,0],""B"":[3,0,0],""C"":[0,0,1],""new_shelf"":0,""violation"":1}","{""A"":[2,3,0],""B"":[3,0,0],""C"":[0,0,1],""new_shelf"":0,""violation"":1}",,,0,3,full_weight_transformers,1337,0,0.6822282075881958,1,1,1,1,1,,,,,,,,,0,0,,,,,,,,1,"Initial A=[2, 2, 0] B=[3, 0, 0] C=[0, 0, 1]
+Arrivals=['P']
+Arrival 1: P
+  Try A: can_place=1 state=[2, 2, 0]
+  Placed on A: new_state=[2, 3, 0]
+Final A=[2, 3, 0] B=[3, 0, 0] C=[0, 0, 1] new_shelf=0 violation=1",1,1,,1,,,1,,,,,,,,,
+20260210_215029,20260210_223824,full_weight_transformers,json_multistep,ms_easy_02,best_of_k,125.48065209388733,0,"{
+  ""plan"": [""A"", ""B""],
+  ""checks"": [
+    {""step"": 1, ""item"": ""S"", ""choice"": ""A"", ""can_place"": 1},
+    {""step"": 2, ""item"": ""P"", ""choice"": ""B"", ""can_place"": 1}
+  ],
+  ""final"": {
+    ""A"": [3, 1, 0],
+    ""B"": [0, 1, 0],
+    ""C"": [0, 0, 0],
+    ""new_shelf"": 0,
+    ""violation"": 0
+  },
+  ""stop"": false,
+  ""stop_reason"": """",
+  ""stop_step_index"": -1
+}",,2,0,1,-1,-1,0,1,1,oracle_equiv_ok,"[""A"",""A""]","[""A"",""B""]","{""A"":[4,1,0],""B"":[0,0,0],""C"":[0,0,0],""new_shelf"":0,""violation"":1}","{""A"":[4,0,0],""B"":[0,1,0],""C"":[0,0,0],""new_shelf"":0,""violation"":1}",,,0,3,full_weight_transformers,1437,0,0.6678733825683594,0,0,1,1,1,,,,,,,,,0,0,,,,,,,,0,"Initial A=[3, 0, 0] B=[0, 0, 0] C=[0, 0, 0]
+Arrivals=['S', 'P']
+Arrival 1: S
+  Try A: can_place=1 state=[3, 0, 0]
+  Placed on A: new_state=[4, 0, 0]
+Arrival 2: P
+  Try A: can_place=1 state=[4, 0, 0]
+  Placed on A: new_state=[4, 1, 0]
+Final A=[4, 1, 0] B=[0, 0, 0] C=[0, 0, 0] new_shelf=0 violation=1",1,1,,1,,,1,,,,,,,,,
+20260210_215029,20260210_223824,full_weight_transformers,json_multistep,ms_med_01,best_of_k,149.3066966533661,0,"{
+  ""plan"": [""A"", ""B"", ""C""],
+  ""checks"": [
+    {""step"": 1, ""item"": ""S"", ""choice"": ""A"", ""can_place"": 1},
+    {""step"": 2, ""item"": ""P"", ""choice"": ""B"", ""can_place"": 1},
+    {""step"": 3, ""item"": ""H"", ""choice"": ""C"", ""can_place"": 1}
+  ],
+  ""final"": {
+    ""A"": [2, 2, 0],
+    ""B"": [0, 0, 0],
+    ""C"": [0, 0, 3],
+    ""new_shelf"": 0,
+    ""violation"": 0
+  },
+  ""stop"": false,
+  ""stop_reason"": """",
+  ""stop_step_index"": -1
+}",,3,1,3,-1,-1,1,1,3,ok,"[""A"",""B"",""C""]","[""A"",""B"",""C""]","{""A"":[3,3,0],""B"":[0,1,0],""C"":[0,0,3],""new_shelf"":0,""violation"":1}","{""A"":[3,3,0],""B"":[0,1,0],""C"":[0,0,3],""new_shelf"":0,""violation"":1}",,,0,3,full_weight_transformers,1537,0,0.6734610795974731,1,1,1,1,1,,,,,,,,,0,0,,,,,,,,1,"Initial A=[2, 3, 0] B=[0, 0, 0] C=[0, 0, 2]
+Arrivals=['S', 'P', 'H']
+Arrival 1: S
+  Try A: can_place=1 state=[2, 3, 0]
+  Placed on A: new_state=[3, 3, 0]
+Arrival 2: P
+  Try A: can_place=0 state=[3, 3, 0]
+  Try B: can_place=1 state=[0, 0, 0]
+  Placed on B: new_state=[0, 1, 0]
+Arrival 3: H
+  Try A: can_place=0 state=[3, 3, 0]
+  Try B: can_place=0 state=[0, 1, 0]
+  Try C: can_place=1 state=[0, 0, 2]
+  Placed on C: new_state=[0, 0, 3]
+Final A=[3, 3, 0] B=[0, 1, 0] C=[0, 0, 3] new_shelf=0 violation=1",1,1,,1,,,1,,,,,,,,,
+20260210_215029,20260210_223824,full_weight_transformers,json_multistep,ms_med_02,best_of_k,144.79933547973633,0,"{
+  ""plan"": [""A"", ""B"", ""C""],
+  ""checks"": [
+    {""step"": 1, ""item"": ""P"", ""choice"": ""A"", ""can_place"": 1},
+    {""step"": 2, ""item"": ""S"", ""choice"": ""B"", ""can_place"": 1},
+    {""step"": 3, ""item"": ""H"", ""choice"": ""C"", ""can_place"": 1}
+  ],
+  ""final"": {
+    ""A"": [4, 0, 0],
+    ""B"": [0, 3, 0],
+    ""C"": [0, 0, 2],
+    ""new_shelf"": 0,
+    ""violation"": 0
+  },
+  ""stop"": false,
+  ""stop_reason"": """",
+  ""stop_step_index"": -1
+}",,3,1,3,-1,-1,1,1,3,ok,"[""A"",""B"",""C""]","[""A"",""B"",""C""]","{""A"":[5,1,0],""B"":[1,4,0],""C"":[0,0,2],""new_shelf"":0,""violation"":1}","{""A"":[5,1,0],""B"":[1,4,0],""C"":[0,0,2],""new_shelf"":0,""violation"":1}",,,0,3,full_weight_transformers,1637,0,0.672205925,1,1,1,1,1,,,,,,,,,0,0,,,,,,,,1,"Initial A=[5, 0, 0] B=[0, 4, 0] C=[0, 0, 1]
+Arrivals=['P', 'S', 'H']
+Arrival 1: P
+  Try A: can_place=1 state=[5, 0, 0]
+  Placed on A: new_state=[5, 1, 0]
+Arrival 2: S
+  Try A: can_place=0 state=[5, 1, 0]
+  Try B: can_place=1 state=[0, 4, 0]
+  Placed on B: new_state=[1, 4, 0]
+Arrival 3: H
+  Try A: can_place=0 state=[5, 1, 0]
+  Try B: can_place=0 state=[1, 4, 0]
+  Try C: can_place=1 state=[0, 0, 1]
+  Placed on C: new_state=[0, 0, 2]
+Final A=[5, 1, 0] B=[1, 4, 0] C=[0, 0, 2] new_shelf=0 violation=1",1,1,,1,,,1,,,,,,,,,
+20260210_215029,20260210_223824,full_weight_transformers,json_multistep,ms_hard_01,best_of_k,137.51127171516418,0,"{
+  ""plan"": [""A"", ""B"", ""C""],
+  ""checks"": [
+    {""step"": 1, ""item"": ""S"", ""choice"": ""A"", ""can_place"": 1},
+    {""step"": 2, ""item"": ""S"", ""choice"": ""B"", ""can_place"": 1},
+    {""step"": 3, ""item"": ""H"", ""choice"": ""C"", ""can_place"": 1}
+  ],
+  ""final"": {
+    ""A"": [1, 5, 0],
+    ""B"": [2, 0, 0],
+    ""C"": [0, 0, 1],
+    ""new_shelf"": 0,
+    ""violation"": 0
+  },
+  ""stop"": false,
+  ""stop_reason"": """",
+  ""stop_step_index"": -1
+}",,3,1,3,-1,-1,1,1,3,ok,"[""A"",""B"",""C""]","[""A"",""B"",""C""]","{""A"":[1,5,0],""B"":[3,0,0],""C"":[0,0,1],""new_shelf"":0,""violation"":1}","{""A"":[1,5,0],""B"":[3,0,0],""C"":[0,0,1],""new_shelf"":0,""violation"":1}",,,0,3,full_weight_transformers,1737,0,0.6760021448135376,1,1,1,1,1,,,,,,,,,0,0,,,,,,,,1,"Initial A=[0, 5, 0] B=[2, 0, 0] C=[0, 0, 0]
+Arrivals=['S', 'S', 'H']
+Arrival 1: S
+  Try A: can_place=1 state=[0, 5, 0]
+  Placed on A: new_state=[1, 5, 0]
+Arrival 2: S
+  Try A: can_place=0 state=[1, 5, 0]
+  Try B: can_place=1 state=[2, 0, 0]
+  Placed on B: new_state=[3, 0, 0]
+Arrival 3: H
+  Try A: can_place=0 state=[1, 5, 0]
+  Try B: can_place=0 state=[3, 0, 0]
+  Try C: can_place=1 state=[0, 0, 0]
+  Placed on C: new_state=[0, 0, 1]
+Final A=[1, 5, 0] B=[3, 0, 0] C=[0, 0, 1] new_shelf=0 violation=1",1,1,,1,,,1,,,,,,,,,
+20260210_215029,20260210_223852,full_weight_transformers,stateful_followup,state_01,2turn,10.44,0,"{""counter"": 2}
+
+{""counter"": 5}",,,,,,,,,2,ok,,,,,,,,,,,,,,,,,,,,,,,,"{""counter"":2}","{""counter"":5}",,,,,,"{""counter"":2}","{""counter"":5}",,,,,,,,,,,,,,,,,1,1,1,1
+20260210_215029,20260210_223852,full_weight_transformers,stateful_followup,state_02,2turn,17.06,0,"{""items"": [""a"", ""b""]}
+
+{""items"": [""a"", ""b"", ""c""]}",,,,,,,,,2,ok,,,,,,,,,,,,,,,,,,,,,,,,"{""items"":[""a"",""b""]}","{""items"":[""a"",""b"",""c""]}",,,,,,"{""items"":[""a"",""b""]}","{""items"":[""a"",""b"",""c""]}",,,,,,,,,,,,,,,,,1,1,1,1
+20260210_215029,20260210_223930,full_weight_transformers,mixed_brief_json,mixed_01,,20.12,0,"ANSWER: 13
+{""a"": 4, ""b"": 9, ""sum"": 13}",,,,,,,,,2,ok,,,,,1,,,,,,,,,,,,,,,,ANSWER: 13,,"{""a"":4,""b"":9,""sum"":13}",,,,,ANSWER: 13,,"{""a"":4,""b"":9,""sum"":13}",,,1,1,,,,,,1,,,,,,,,,,,,
+20260210_215029,20260210_223930,full_weight_transformers,mixed_brief_json,mixed_02,,18.67,0,"ANSWER: 6
+{""a"": -2, ""b"": 8, ""sum"": 6}",,,,,,,,,2,ok,,,,,1,,,,,,,,,,,,,,,,ANSWER: 6,,"{""a"":-2,""b"":8,""sum"":6}",,,,,ANSWER: 6,,"{""a"":-2,""b"":8,""sum"":6}",,,1,1,,,,,,1,,,,,,,,,,,,
+20260210_215029,20260210_224116,full_weight_transformers,fuzz,fuzz_0000,tool_decisions,105.31,0,,,5,1,5,-1,-1,1,1,10,ok,"[""A"",""B"",""B"",""B"",""B""]","[""A"",""B"",""B"",""B"",""B""]","{""A"":[0,6,0],""B"":[5,1,0],""C"":[5,0,0],""new_shelf"":0,""violation"":1}","{""A"":[0,6,0],""B"":[5,1,0],""C"":[5,0,0],""new_shelf"":0,""violation"":1}",,,,,,,,,,,,,,,ok,,,,,,,,,,,,,,,,,"Initial A=[0, 5, 0] B=[1, 1, 0] C=[5, 0, 0]
+Arrivals=['P', 'S', 'S', 'S', 'S']
+Arrival 1: P
+  Try A: can_place=1 state=[0, 5, 0]
+  Placed on A: new_state=[0, 6, 0]
+Arrival 2: S
+  Try A: can_place=0 state=[0, 6, 0]
+  Try B: can_place=1 state=[1, 1, 0]
+  Placed on B: new_state=[2, 1, 0]
+Arrival 3: S
+  Try A: can_place=0 state=[0, 6, 0]
+  Try B: can_place=1 state=[2, 1, 0]
+  Placed on B: new_state=[3, 1, 0]
+Arrival 4: S
+  Try A: can_place=0 state=[0, 6, 0]
+  Try B: can_place=1 state=[3, 1, 0]
+  Placed on B: new_state=[4, 1, 0]
+Arrival 5: S
+  Try A: can_place=0 state=[0, 6, 0]
+  Try B: can_place=1 state=[4, 1, 0]
+  Placed on B: new_state=[5, 1, 0]
+Final A=[0, 6, 0] B=[5, 1, 0] C=[5, 0, 0] new_shelf=0 violation=1",,,,,,,,,,,,,,,,
+20260210_215029,20260210_224137,full_weight_transformers,fuzz,fuzz_0001,tool_decisions,21.13,0,,,1,1,1,-1,-1,1,1,10,ok,"[""A""]","[""A""]","{""A"":[5,0,0],""B"":[5,1,0],""C"":[4,1,0],""new_shelf"":0,""violation"":1}","{""A"":[5,0,0],""B"":[5,1,0],""C"":[4,1,0],""new_shelf"":0,""violation"":1}",,,,,,,,,,,,,,,ok,,,,,,,,,,,,,,,,,"Initial A=[4, 0, 0] B=[5, 1, 0] C=[4, 1, 0]
+Arrivals=['S']
+Arrival 1: S
+  Try A: can_place=1 state=[4, 0, 0]
+  Placed on A: new_state=[5, 0, 0]
+Final A=[5, 0, 0] B=[5, 1, 0] C=[4, 1, 0] new_shelf=0 violation=1",,,,,,,,,,,,,,,,
+20260210_215029,20260210_224156,full_weight_transformers,fuzz,fuzz_0002,tool_decisions,18.96,0,,,1,1,1,-1,-1,1,1,10,ok,"[""A""]","[""A""]","{""A"":[2,3,0],""B"":[0,0,6],""C"":[0,3,0],""new_shelf"":0,""violation"":1}","{""A"":[2,3,0],""B"":[0,0,6],""C"":[0,3,0],""new_shelf"":0,""violation"":1}",,,,,,,,,,,,,,,ok,,,,,,,,,,,,,,,,,"Initial A=[2, 2, 0] B=[0, 0, 6] C=[0, 3, 0]
+Arrivals=['P']
+Arrival 1: P
+  Try A: can_place=1 state=[2, 2, 0]
+  Placed on A: new_state=[2, 3, 0]
+Final A=[2, 3, 0] B=[0, 0, 6] C=[0, 3, 0] new_shelf=0 violation=1",,,,,,,,,,,,,,,,
+20260210_215029,20260210_224239,full_weight_transformers,fuzz,fuzz_0003,tool_decisions,43.09,0,,,5,1,5,1,1,1,1,10,ok,"[""A"",""STOP"",""STOP"",""STOP"",""STOP""]","[""A"",""STOP"",""STOP"",""STOP"",""STOP""]","{""A"":[4,2,0],""B"":[5,1,0],""C"":[3,0,0],""new_shelf"":1,""violation"":1}","{""A"":[4,2,0],""B"":[5,1,0],""C"":[3,0,0],""new_shelf"":1,""violation"":1}",,,,,,,,,,,,,,,ok,,,,,,,,,,,,,,,,,"Initial A=[4, 1, 0] B=[5, 1, 0] C=[3, 0, 0]
+Arrivals=['P', 'H', 'H', 'P', 'H']
+Arrival 1: P
+  Try A: can_place=1 state=[4, 1, 0]
+  Placed on A: new_state=[4, 2, 0]
+Arrival 2: H
+  Try A: can_place=0 state=[4, 2, 0]
+  Try B: can_place=0 state=[5, 1, 0]
+  Try C: can_place=0 state=[3, 0, 0]
+  No placement possible on A/B/C: placements=STOP
+Arrival 3: H
+  Already stopped: placements=STOP
+Arrival 4: P
+  Already stopped: placements=STOP
+Arrival 5: H
+  Already stopped: placements=STOP
+Final A=[4, 2, 0] B=[5, 1, 0] C=[3, 0, 0] new_shelf=1 violation=1",,,,,,,,,,,,,,,,
+20260210_215029,20260210_224322,full_weight_transformers,fuzz,fuzz_0004,tool_decisions,42.64,0,,,2,1,2,-1,-1,1,1,10,ok,"[""A"",""A""]","[""A"",""A""]","{""A"":[1,1,0],""B"":[0,0,1],""C"":[0,6,0],""new_shelf"":0,""violation"":1}","{""A"":[1,1,0],""B"":[0,0,1],""C"":[0,6,0],""new_shelf"":0,""violation"":1}",,,,,,,,,,,,,,,ok,,,,,,,,,,,,,,,,,"Initial A=[0, 0, 0] B=[0, 0, 1] C=[0, 6, 0]
+Arrivals=['S', 'P']
+Arrival 1: S
+  Try A: can_place=1 state=[0, 0, 0]
+  Placed on A: new_state=[1, 0, 0]
+Arrival 2: P
+  Try A: can_place=1 state=[1, 0, 0]
+  Placed on A: new_state=[1, 1, 0]
+Final A=[1, 1, 0] B=[0, 0, 1] C=[0, 6, 0] new_shelf=0 violation=1",,,,,,,,,,,,,,,,
+20260210_215029,20260210_224459,full_weight_transformers,fuzz,fuzz_0005,tool_decisions,96.96,0,,,5,1,5,-1,-1,1,1,10,ok,"[""B"",""C"",""B"",""B"",""C""]","[""B"",""C"",""B"",""B"",""C""]","{""A"":[5,1,0],""B"":[0,0,5],""C"":[4,0,0],""new_shelf"":0,""violation"":1}","{""A"":[5,1,0],""B"":[0,0,5],""C"":[4,0,0],""new_shelf"":0,""violation"":1}",,,,,,,,,,,,,,,ok,,,,,,,,,,,,,,,,,"Initial A=[5, 1, 0] B=[0, 0, 2] C=[2, 0, 0]
+Arrivals=['H', 'S', 'H', 'H', 'S']
+Arrival 1: H
+  Try A: can_place=0 state=[5, 1, 0]
+  Try B: can_place=1 state=[0, 0, 2]
+  Placed on B: new_state=[0, 0, 3]
+Arrival 2: S
+  Try A: can_place=0 state=[5, 1, 0]
+  Try B: can_place=0 state=[0, 0, 3]
+  Try C: can_place=1 state=[2, 0, 0]
+  Placed on C: new_state=[3, 0, 0]
+Arrival 3: H
+  Try A: can_place=0 state=[5, 1, 0]
+  Try B: can_place=1 state=[0, 0, 3]
+  Placed on B: new_state=[0, 0, 4]
+Arrival 4: H
+  Try A: can_place=0 state=[5, 1, 0]
+  Try B: can_place=1 state=[0, 0, 4]
+  Placed on B: new_state=[0, 0, 5]
+Arrival 5: S
+  Try A: can_place=0 state=[5, 1, 0]
+  Try B: can_place=0 state=[0, 0, 5]
+  Try C: can_place=1 state=[3, 0, 0]
+  Placed on C: new_state=[4, 0, 0]
+Final A=[5, 1, 0] B=[0, 0, 5] C=[4, 0, 0] new_shelf=0 violation=1",,,,,,,,,,,,,,,,
+20260210_215029,20260210_224538,full_weight_transformers,fuzz,fuzz_0006,tool_decisions,39.27,0,,,2,1,2,1,1,1,1,10,ok,"[""B"",""STOP""]","[""B"",""STOP""]","{""A"":[0,0,3],""B"":[6,0,0],""C"":[6,0,0],""new_shelf"":1,""violation"":0}","{""A"":[0,0,3],""B"":[6,0,0],""C"":[6,0,0],""new_shelf"":1,""violation"":0}",,,,,,,,,,,,,,,ok,,,,,,,,,,,,,,,,,"Initial A=[0, 0, 3] B=[5, 0, 0] C=[6, 0, 0]
+Arrivals=['S', 'P']
+Arrival 1: S
+  Try A: can_place=0 state=[0, 0, 3]
+  Try B: can_place=1 state=[5, 0, 0]
+  Placed on B: new_state=[6, 0, 0]
+Arrival 2: P
+  Try A: can_place=0 state=[0, 0, 3]
+  Try B: can_place=0 state=[6, 0, 0]
+  Try C: can_place=0 state=[6, 0, 0]
+  No placement possible on A/B/C: placements=STOP
+Final A=[0, 0, 3] B=[6, 0, 0] C=[6, 0, 0] new_shelf=1 violation=0",,,,,,,,,,,,,,,,
+20260210_215029,20260210_224557,full_weight_transformers,fuzz,fuzz_0007,tool_decisions,18.92,0,,,4,1,4,0,0,1,1,10,ok,"[""STOP"",""STOP"",""STOP"",""STOP""]","[""STOP"",""STOP"",""STOP"",""STOP""]","{""A"":[0,1,0],""B"":[4,2,0],""C"":[5,1,0],""new_shelf"":1,""violation"":1}","{""A"":[0,1,0],""B"":[4,2,0],""C"":[5,1,0],""new_shelf"":1,""violation"":1}",,,,,,,,,,,,,,,ok,,,,,,,,,,,,,,,,,"Initial A=[0, 1, 0] B=[4, 2, 0] C=[5, 1, 0]
+Arrivals=['H', 'P', 'S', 'P']
+Arrival 1: H
+  Try A: can_place=0 state=[0, 1, 0]
+  Try B: can_place=0 state=[4, 2, 0]
+  Try C: can_place=0 state=[5, 1, 0]
+  No placement possible on A/B/C: placements=STOP
+Arrival 2: P
+  Already stopped: placements=STOP
+Arrival 3: S
+  Already stopped: placements=STOP
+Arrival 4: P
+  Already stopped: placements=STOP
+Final A=[0, 1, 0] B=[4, 2, 0] C=[5, 1, 0] new_shelf=1 violation=1",,,,,,,,,,,,,,,,
+20260210_215029,20260210_224641,full_weight_transformers,fuzz,fuzz_0008,tool_decisions,44.42,0,,,3,1,3,1,1,1,1,10,ok,"[""C"",""STOP"",""STOP""]","[""C"",""STOP"",""STOP""]","{""A"":[0,0,5],""B"":[2,4,0],""C"":[5,1,0],""new_shelf"":1,""violation"":0}","{""A"":[0,0,5],""B"":[2,4,0],""C"":[5,1,0],""new_shelf"":1,""violation"":0}",,,,,,,,,,,,,,,ok,,,,,,,,,,,,,,,,,"Initial A=[0, 0, 5] B=[2, 4, 0] C=[4, 1, 0]
+Arrivals=['S', 'S', 'H']
+Arrival 1: S
+  Try A: can_place=0 state=[0, 0, 5]
+  Try B: can_place=0 state=[2, 4, 0]
+  Try C: can_place=1 state=[4, 1, 0]
+  Placed on C: new_state=[5, 1, 0]
+Arrival 2: S
+  Try A: can_place=0 state=[0, 0, 5]
+  Try B: can_place=0 state=[2, 4, 0]
+  Try C: can_place=0 state=[5, 1, 0]
+  No placement possible on A/B/C: placements=STOP
+Arrival 3: H
+  Already stopped: placements=STOP
+Final A=[0, 0, 5] B=[2, 4, 0] C=[5, 1, 0] new_shelf=1 violation=0",,,,,,,,,,,,,,,,
+20260210_215029,20260210_224725,full_weight_transformers,fuzz,fuzz_0009,tool_decisions,44.1,0,,,4,1,4,1,1,1,1,10,ok,"[""C"",""STOP"",""STOP"",""STOP""]","[""C"",""STOP"",""STOP"",""STOP""]","{""A"":[6,0,0],""B"":[1,5,0],""C"":[0,0,6],""new_shelf"":1,""violation"":1}","{""A"":[6,0,0],""B"":[1,5,0],""C"":[0,0,6],""new_shelf"":1,""violation"":1}",,,,,,,,,,,,,,,ok,,,,,,,,,,,,,,,,,"Initial A=[6, 0, 0] B=[1, 5, 0] C=[0, 0, 5]
+Arrivals=['H', 'S', 'P', 'P']
+Arrival 1: H
+  Try A: can_place=0 state=[6, 0, 0]
+  Try B: can_place=0 state=[1, 5, 0]
+  Try C: can_place=1 state=[0, 0, 5]
+  Placed on C: new_state=[0, 0, 6]
+Arrival 2: S
+  Try A: can_place=0 state=[6, 0, 0]
+  Try B: can_place=0 state=[1, 5, 0]
+  Try C: can_place=0 state=[0, 0, 6]
+  No placement possible on A/B/C: placements=STOP
+Arrival 3: P
+  Already stopped: placements=STOP
+Arrival 4: P
+  Already stopped: placements=STOP
+Final A=[6, 0, 0] B=[1, 5, 0] C=[0, 0, 6] new_shelf=1 violation=1",,,,,,,,,,,,,,,,
+20260210_215029,20260210_224805,full_weight_transformers,fuzz,fuzz_0010,tool_decisions,39.67,0,,,5,1,5,1,1,1,1,10,ok,"[""A"",""STOP"",""STOP"",""STOP"",""STOP""]","[""A"",""STOP"",""STOP"",""STOP"",""STOP""]","{""A"":[3,3,0],""B"":[6,0,0],""C"":[0,5,0],""new_shelf"":1,""violation"":1}","{""A"":[3,3,0],""B"":[6,0,0],""C"":[0,5,0],""new_shelf"":1,""violation"":1}",,,,,,,,,,,,,,,ok,,,,,,,,,,,,,,,,,"Initial A=[3, 2, 0] B=[6, 0, 0] C=[0, 5, 0]
+Arrivals=['P', 'H', 'P', 'S', 'P']
+Arrival 1: P
+  Try A: can_place=1 state=[3, 2, 0]
+  Placed on A: new_state=[3, 3, 0]
+Arrival 2: H
+  Try A: can_place=0 state=[3, 3, 0]
+  Try B: can_place=0 state=[6, 0, 0]
+  Try C: can_place=0 state=[0, 5, 0]
+  No placement possible on A/B/C: placements=STOP
+Arrival 3: P
+  Already stopped: placements=STOP
+Arrival 4: S
+  Already stopped: placements=STOP
+Arrival 5: P
+  Already stopped: placements=STOP
+Final A=[3, 3, 0] B=[6, 0, 0] C=[0, 5, 0] new_shelf=1 violation=1",,,,,,,,,,,,,,,,
+20260210_215029,20260210_224847,full_weight_transformers,fuzz,fuzz_0011,tool_decisions,41.56,0,,,5,1,5,1,1,1,1,10,ok,"[""A"",""STOP"",""STOP"",""STOP"",""STOP""]","[""A"",""STOP"",""STOP"",""STOP"",""STOP""]","{""A"":[4,0,0],""B"":[5,1,0],""C"":[6,0,0],""new_shelf"":1,""violation"":1}","{""A"":[4,0,0],""B"":[5,1,0],""C"":[6,0,0],""new_shelf"":1,""violation"":1}",,,,,,,,,,,,,,,ok,,,,,,,,,,,,,,,,,"Initial A=[3, 0, 0] B=[5, 1, 0] C=[6, 0, 0]
+Arrivals=['S', 'H', 'P', 'H', 'H']
+Arrival 1: S
+  Try A: can_place=1 state=[3, 0, 0]
+  Placed on A: new_state=[4, 0, 0]
+Arrival 2: H
+  Try A: can_place=0 state=[4, 0, 0]
+  Try B: can_place=0 state=[5, 1, 0]
+  Try C: can_place=0 state=[6, 0, 0]
+  No placement possible on A/B/C: placements=STOP
+Arrival 3: P
+  Already stopped: placements=STOP
+Arrival 4: H
+  Already stopped: placements=STOP
+Arrival 5: H
+  Already stopped: placements=STOP
+Final A=[4, 0, 0] B=[5, 1, 0] C=[6, 0, 0] new_shelf=1 violation=1",,,,,,,,,,,,,,,,
+20260210_215029,20260210_225007,full_weight_transformers,fuzz,fuzz_0012,tool_decisions,80.1,0,,,4,1,4,-1,-1,1,1,10,ok,"[""A"",""A"",""A"",""C""]","[""A"",""A"",""A"",""C""]","{""A"":[3,3,0],""B"":[4,2,0],""C"":[4,2,0],""new_shelf"":0,""violation"":1}","{""A"":[3,3,0],""B"":[4,2,0],""C"":[4,2,0],""new_shelf"":0,""violation"":1}",,,,,,,,,,,,,,,ok,,,,,,,,,,,,,,,,,"Initial A=[1, 2, 0] B=[4, 2, 0] C=[4, 1, 0]
+Arrivals=['S', 'S', 'P', 'P']
+Arrival 1: S
+  Try A: can_place=1 state=[1, 2, 0]
+  Placed on A: new_state=[2, 2, 0]
+Arrival 2: S
+  Try A: can_place=1 state=[2, 2, 0]
+  Placed on A: new_state=[3, 2, 0]
+Arrival 3: P
+  Try A: can_place=1 state=[3, 2, 0]
+  Placed on A: new_state=[3, 3, 0]
+Arrival 4: P
+  Try A: can_place=0 state=[3, 3, 0]
+  Try B: can_place=0 state=[4, 2, 0]
+  Try C: can_place=1 state=[4, 1, 0]
+  Placed on C: new_state=[4, 2, 0]
+Final A=[3, 3, 0] B=[4, 2, 0] C=[4, 2, 0] new_shelf=0 violation=1",,,,,,,,,,,,,,,,
+20260210_215029,20260210_225105,full_weight_transformers,fuzz,fuzz_0013,tool_decisions,58.21,0,,,5,1,5,2,2,1,1,10,ok,"[""A"",""B"",""STOP"",""STOP"",""STOP""]","[""A"",""B"",""STOP"",""STOP"",""STOP""]","{""A"":[2,4,0],""B"":[1,5,0],""C"":[0,6,0],""new_shelf"":1,""violation"":1}","{""A"":[2,4,0],""B"":[1,5,0],""C"":[0,6,0],""new_shelf"":1,""violation"":1}",,,,,,,,,,,,,,,ok,,,,,,,,,,,,,,,,,"Initial A=[1, 4, 0] B=[0, 5, 0] C=[0, 6, 0]
+Arrivals=['S', 'S', 'H', 'P', 'H']
+Arrival 1: S
+  Try A: can_place=1 state=[1, 4, 0]
+  Placed on A: new_state=[2, 4, 0]
+Arrival 2: S
+  Try A: can_place=0 state=[2, 4, 0]
+  Try B: can_place=1 state=[0, 5, 0]
+  Placed on B: new_state=[1, 5, 0]
+Arrival 3: H
+  Try A: can_place=0 state=[2, 4, 0]
+  Try B: can_place=0 state=[1, 5, 0]
+  Try C: can_place=0 state=[0, 6, 0]
+  No placement possible on A/B/C: placements=STOP
+Arrival 4: P
+  Already stopped: placements=STOP
+Arrival 5: H
+  Already stopped: placements=STOP
+Final A=[2, 4, 0] B=[1, 5, 0] C=[0, 6, 0] new_shelf=1 violation=1",,,,,,,,,,,,,,,,
+20260210_215029,20260210_225247,full_weight_transformers,fuzz,fuzz_0014,tool_decisions,102.3,0,,,6,1,6,4,4,1,1,10,ok,"[""B"",""B"",""A"",""A"",""STOP"",""STOP""]","[""B"",""B"",""A"",""A"",""STOP"",""STOP""]","{""A"":[0,0,6],""B"":[3,3,0],""C"":[6,0,0],""new_shelf"":1,""violation"":1}","{""A"":[0,0,6],""B"":[3,3,0],""C"":[6,0,0],""new_shelf"":1,""violation"":1}",,,,,,,,,,,,,,,ok,,,,,,,,,,,,,,,,,"Initial A=[0, 0, 4] B=[3, 1, 0] C=[6, 0, 0]
+Arrivals=['P', 'P', 'H', 'H', 'P', 'P']
+Arrival 1: P
+  Try A: can_place=0 state=[0, 0, 4]
+  Try B: can_place=1 state=[3, 1, 0]
+  Placed on B: new_state=[3, 2, 0]
+Arrival 2: P
+  Try A: can_place=0 state=[0, 0, 4]
+  Try B: can_place=1 state=[3, 2, 0]
+  Placed on B: new_state=[3, 3, 0]
+Arrival 3: H
+  Try A: can_place=1 state=[0, 0, 4]
+  Placed on A: new_state=[0, 0, 5]
+Arrival 4: H
+  Try A: can_place=1 state=[0, 0, 5]
+  Placed on A: new_state=[0, 0, 6]
+Arrival 5: P
+  Try A: can_place=0 state=[0, 0, 6]
+  Try B: can_place=0 state=[3, 3, 0]
+  Try C: can_place=0 state=[6, 0, 0]
+  No placement possible on A/B/C: placements=STOP
+Arrival 6: P
+  Already stopped: placements=STOP
+Final A=[0, 0, 6] B=[3, 3, 0] C=[6, 0, 0] new_shelf=1 violation=1",,,,,,,,,,,,,,,,
+20260210_215029,20260210_225331,full_weight_transformers,fuzz,fuzz_0015,tool_decisions,44.1,0,,,2,1,2,-1,-1,1,1,10,ok,"[""C"",""A""]","[""C"",""A""]","{""A"":[4,0,0],""B"":[0,2,0],""C"":[0,0,5],""new_shelf"":0,""violation"":1}","{""A"":[4,0,0],""B"":[0,2,0],""C"":[0,0,5],""new_shelf"":0,""violation"":1}",,,,,,,,,,,,,,,ok,,,,,,,,,,,,,,,,,"Initial A=[3, 0, 0] B=[0, 2, 0] C=[0, 0, 4]
+Arrivals=['H', 'S']
+Arrival 1: H
+  Try A: can_place=0 state=[3, 0, 0]
+  Try B: can_place=0 state=[0, 2, 0]
+  Try C: can_place=1 state=[0, 0, 4]
+  Placed on C: new_state=[0, 0, 5]
+Arrival 2: S
+  Try A: can_place=1 state=[3, 0, 0]
+  Placed on A: new_state=[4, 0, 0]
+Final A=[4, 0, 0] B=[0, 2, 0] C=[0, 0, 5] new_shelf=0 violation=1",,,,,,,,,,,,,,,,
+20260210_215029,20260210_225350,full_weight_transformers,fuzz,fuzz_0016,tool_decisions,19.02,0,,,1,1,1,-1,-1,1,1,10,ok,"[""A""]","[""A""]","{""A"":[0,0,6],""B"":[1,0,0],""C"":[6,0,0],""new_shelf"":0,""violation"":1}","{""A"":[0,0,6],""B"":[1,0,0],""C"":[6,0,0],""new_shelf"":0,""violation"":1}",,,,,,,,,,,,,,,ok,,,,,,,,,,,,,,,,,"Initial A=[0, 0, 5] B=[1, 0, 0] C=[6, 0, 0]
+Arrivals=['H']
+Arrival 1: H
+  Try A: can_place=1 state=[0, 0, 5]
+  Placed on A: new_state=[0, 0, 6]
+Final A=[0, 0, 6] B=[1, 0, 0] C=[6, 0, 0] new_shelf=0 violation=1",,,,,,,,,,,,,,,,
+20260210_215029,20260210_225411,full_weight_transformers,fuzz,fuzz_0017,tool_decisions,20.66,0,,,4,1,4,0,0,1,1,10,ok,"[""STOP"",""STOP"",""STOP"",""STOP""]","[""STOP"",""STOP"",""STOP"",""STOP""]","{""A"":[5,1,0],""B"":[1,5,0],""C"":[4,2,0],""new_shelf"":1,""violation"":1}","{""A"":[5,1,0],""B"":[1,5,0],""C"":[4,2,0],""new_shelf"":1,""violation"":1}",,,,,,,,,,,,,,,ok,,,,,,,,,,,,,,,,,"Initial A=[5, 1, 0] B=[1, 5, 0] C=[4, 2, 0]
+Arrivals=['S', 'P', 'P', 'S']
+Arrival 1: S
+  Try A: can_place=0 state=[5, 1, 0]
+  Try B: can_place=0 state=[1, 5, 0]
+  Try C: can_place=0 state=[4, 2, 0]
+  No placement possible on A/B/C: placements=STOP
+Arrival 2: P
+  Already stopped: placements=STOP
+Arrival 3: P
+  Already stopped: placements=STOP
+Arrival 4: S
+  Already stopped: placements=STOP
+Final A=[5, 1, 0] B=[1, 5, 0] C=[4, 2, 0] new_shelf=1 violation=1",,,,,,,,,,,,,,,,
+20260210_215029,20260210_225454,full_weight_transformers,fuzz,fuzz_0018,tool_decisions,43.09,0,,,6,1,6,1,1,1,1,10,ok,"[""A"",""STOP"",""STOP"",""STOP"",""STOP"",""STOP""]","[""A"",""STOP"",""STOP"",""STOP"",""STOP"",""STOP""]","{""A"":[0,0,6],""B"":[3,3,0],""C"":[5,0,0],""new_shelf"":1,""violation"":1}","{""A"":[0,0,6],""B"":[3,3,0],""C"":[5,0,0],""new_shelf"":1,""violation"":1}",,,,,,,,,,,,,,,ok,,,,,,,,,,,,,,,,,"Initial A=[0, 0, 5] B=[3, 3, 0] C=[5, 0, 0]
+Arrivals=['H', 'H', 'S', 'S', 'P', 'H']
+Arrival 1: H
+  Try A: can_place=1 state=[0, 0, 5]
+  Placed on A: new_state=[0, 0, 6]
+Arrival 2: H
+  Try A: can_place=0 state=[0, 0, 6]
+  Try B: can_place=0 state=[3, 3, 0]
+  Try C: can_place=0 state=[5, 0, 0]
+  No placement possible on A/B/C: placements=STOP
+Arrival 3: S
+  Already stopped: placements=STOP
+Arrival 4: S
+  Already stopped: placements=STOP
+Arrival 5: P
+  Already stopped: placements=STOP
+Arrival 6: H
+  Already stopped: placements=STOP
+Final A=[0, 0, 6] B=[3, 3, 0] C=[5, 0, 0] new_shelf=1 violation=1",,,,,,,,,,,,,,,,
+20260210_215029,20260210_225555,full_weight_transformers,fuzz,fuzz_0019,tool_decisions,60.58,0,,,3,1,3,-1,-1,1,1,10,ok,"[""B"",""C"",""C""]","[""B"",""C"",""C""]","{""A"":[6,0,0],""B"":[1,5,0],""C"":[5,1,0],""new_shelf"":0,""violation"":1}","{""A"":[6,0,0],""B"":[1,5,0],""C"":[5,1,0],""new_shelf"":0,""violation"":1}",,,,,,,,,,,,,,,ok,,,,,,,,,,,,,,,,,"Initial A=[6, 0, 0] B=[1, 4, 0] C=[3, 1, 0]
+Arrivals=['P', 'S', 'S']
+Arrival 1: P
+  Try A: can_place=0 state=[6, 0, 0]
+  Try B: can_place=1 state=[1, 4, 0]
+  Placed on B: new_state=[1, 5, 0]
+Arrival 2: S
+  Try A: can_place=0 state=[6, 0, 0]
+  Try B: can_place=0 state=[1, 5, 0]
+  Try C: can_place=1 state=[3, 1, 0]
+  Placed on C: new_state=[4, 1, 0]
+Arrival 3: S
+  Try A: can_place=0 state=[6, 0, 0]
+  Try B: can_place=0 state=[1, 5, 0]
+  Try C: can_place=1 state=[4, 1, 0]
+  Placed on C: new_state=[5, 1, 0]
+Final A=[6, 0, 0] B=[1, 5, 0] C=[5, 1, 0] new_shelf=0 violation=1",,,,,,,,,,,,,,,,
+20260210_215029,20260210_225559,full_weight_transformers,mcq,mcq_01,,0.73,0,B,,,1,,,,1,,1,ok,B,B,,,,,,,,,,,,,,,,B,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
+20260210_215029,20260210_225559,full_weight_transformers,mcq,mcq_02,,0.73,0,B,,,1,,,,1,,1,ok,B,B,,,,,,,,,,,,,,,,B,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
+20260210_215029,20260210_225559,full_weight_transformers,mcq,mcq_03,,0.73,0,C,,,1,,,,1,,1,ok,C,C,,,,,,,,,,,,,,,,C,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
+20260210_215029,20260210_225559,full_weight_transformers,mcq,mcq_04,,0.74,0,B,,,1,,,,1,,1,ok,B,B,,,,,,,,,,,,,,,,B,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
+20260210_215029,20260210_225559,full_weight_transformers,mcq,mcq_05,,0.73,0,B,,,1,,,,1,,1,ok,B,B,,,,,,,,,,,,,,,,B,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
+20260210_215029,20260210_225651,full_weight_transformers,toolcall,tool_01,,26.1,0,"{
+  ""tool_name"": ""add"",
+  ""args"": {
+    ""a"": 2,
+    ""b"": 3
+  }
+}
+
+5",,,,,,,1,,11,ok,5,5,,,,,,,,,,,,,,,,,,ok,,,,,,,,,,,,,,,,,,,,,1,1,,ok,1,,"{""tool_name"": ""add"", ""args"": {""a"": 2, ""b"": 3}}","{
+  ""tool_name"": ""add"",
+  ""args"": {
+    ""a"": 2,
+    ""b"": 3
+  }
+}",,,,
+20260210_215029,20260210_225651,full_weight_transformers,toolcall,tool_02,,25.8,0,"{
+  ""tool_name"": ""add"",
+  ""args"": {
+    ""a"": 10,
+    ""b"": -4
+  }
+}
+
+6",,,,,,,1,,11,ok,6,6,,,,,,,,,,,,,,,,,,ok,,,,,,,,,,,,,,,,,,,,,1,1,,ok,1,,"{""tool_name"": ""add"", ""args"": {""a"": 10, ""b"": -4}}","{
+  ""tool_name"": ""add"",
+  ""args"": {
+    ""a"": 10,
+    ""b"": -4
+  }
+}",,,,
+20260210_215029,20260210_225721,full_weight_transformers,toolcall_only,toolonly_01,,14.95,0,"{""tool"": ""add"", ""left"": 5, ""right"": 10}",,,,,,,,,1,schema_error,,,,,,0,,,,,,,,,,,,,,,,"{""a"":5,""b"":10}",,,,,,,{},,,,,,,,1,,"'args' is a required property; Additional properties are not allowed ('left', 'right' were unexpected)",0,,,,,,1,,,,,,
+20260210_215029,20260210_225721,full_weight_transformers,toolcall_only,toolonly_02,,15.76,0,"{""tool"": ""add"", ""left"": 25, ""right"": 75}",,,,,,,,,1,schema_error,,,,,,0,,,,,,,,,,,,,,,,"{""a"":25,""b"":75}",,,,,,,{},,,,,,,,1,,"'args' is a required property; Additional properties are not allowed ('left', 'right' were unexpected)",0,,,,,,1,,,,,,
--- a/qwen-2.5-14B-instruct-1m-gguf-F16.gguf
+++ b/qwen-2.5-14B-instruct-1m-gguf-F16.gguf
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:de08ea9c41234ef83b7aacf07f9ebc3cbaa20ca8aeb5f6417758a8798660aaa9
+size 29547716384
--- a/rollup.json
+++ b/rollup.json
@@ -0,0 +1,267 @@
+{
+  "by_runner_family": {
+    "full_weight_transformers": {
+      "json_multistep": {
+        "n": 5,
+        "pass_rate": 0.8,
+        "pass_cols": [
+          "schema_ok",
+          "checks_consistent_ok",
+          "stop_semantics_ok",
+          "oracle_equiv_ok"
+        ],
+        "averages": {
+          "bucket_score": 2.2
+        },
+        "top_failures": [
+          {
+            "reason": "oracle_equiv_ok",
+            "count": 1
+          }
+        ],
+        "signal_rates": {
+          "schema_ok": 1.0,
+          "checks_consistent_ok": 1.0,
+          "stop_semantics_ok": 1.0,
+          "oracle_equiv_ok": 0.8
+        },
+        "signal_counts": {
+          "schema_ok": {
+            "present": 5,
+            "ok": 5,
+            "fail": 0
+          },
+          "checks_consistent_ok": {
+            "present": 5,
+            "ok": 5,
+            "fail": 0
+          },
+          "stop_semantics_ok": {
+            "present": 5,
+            "ok": 5,
+            "fail": 0
+          },
+          "oracle_equiv_ok": {
+            "present": 5,
+            "ok": 4,
+            "fail": 1
+          }
+        },
+        "tier2_rates": {
+          "final_consistent_ok": 0.0,
+          "final_match_reported": 0.0
+        },
+        "tier2_counts": {
+          "final_consistent_ok": {
+            "present": 5,
+            "ok": 0,
+            "fail": 5
+          },
+          "final_match_reported": {
+            "present": 5,
+            "ok": 0,
+            "fail": 5
+          }
+        }
+      },
+      "stateful_followup": {
+        "n": 2,
+        "pass_rate": 1.0,
+        "pass_cols": [
+          "turn1_parse_ok",
+          "turn2_parse_ok",
+          "turn1_exact_match",
+          "turn2_exact_match"
+        ],
+        "averages": {
+          "bucket_score": 2.0
+        },
+        "top_failures": [],
+        "signal_rates": {
+          "turn1_parse_ok": 1.0,
+          "turn2_parse_ok": 1.0,
+          "turn1_exact_match": 1.0,
+          "turn2_exact_match": 1.0
+        },
+        "signal_counts": {
+          "turn1_parse_ok": {
+            "present": 2,
+            "ok": 2,
+            "fail": 0
+          },
+          "turn2_parse_ok": {
+            "present": 2,
+            "ok": 2,
+            "fail": 0
+          },
+          "turn1_exact_match": {
+            "present": 2,
+            "ok": 2,
+            "fail": 0
+          },
+          "turn2_exact_match": {
+            "present": 2,
+            "ok": 2,
+            "fail": 0
+          }
+        },
+        "tier2_rates": {},
+        "tier2_counts": {}
+      },
+      "toolcall_only": {
+        "n": 2,
+        "pass_rate": 0.0,
+        "pass_cols": [
+          "tool_name_ok",
+          "args_ok"
+        ],
+        "averages": {
+          "bucket_score": 1.0
+        },
+        "top_failures": [
+          {
+            "reason": "args_ok",
+            "count": 2
+          }
+        ],
+        "signal_rates": {
+          "tool_name_ok": 1.0,
+          "args_ok": 0.0
+        },
+        "signal_counts": {
+          "tool_name_ok": {
+            "present": 2,
+            "ok": 2,
+            "fail": 0
+          },
+          "args_ok": {
+            "present": 2,
+            "ok": 0,
+            "fail": 2
+          }
+        },
+        "tier2_rates": {},
+        "tier2_counts": {}
+      },
+      "mixed_brief_json": {
+        "n": 2,
+        "pass_rate": 1.0,
+        "pass_cols": [
+          "answer_line_ok",
+          "json_parse_ok",
+          "schema_ok"
+        ],
+        "averages": {
+          "bucket_score": 2.0
+        },
+        "top_failures": [],
+        "signal_rates": {
+          "answer_line_ok": 1.0,
+          "json_parse_ok": 1.0,
+          "schema_ok": 1.0
+        },
+        "signal_counts": {
+          "answer_line_ok": {
+            "present": 2,
+            "ok": 2,
+            "fail": 0
+          },
+          "json_parse_ok": {
+            "present": 2,
+            "ok": 2,
+            "fail": 0
+          },
+          "schema_ok": {
+            "present": 2,
+            "ok": 2,
+            "fail": 0
+          }
+        },
+        "tier2_rates": {},
+        "tier2_counts": {}
+      },
+      "json": {
+        "n": 4,
+        "pass_rate": null,
+        "pass_cols": [
+          "schema_ok",
+          "constraints_ok"
+        ],
+        "averages": {
+          "bucket_score": 10.0
+        },
+        "top_failures": [],
+        "signal_rates": {},
+        "signal_counts": {},
+        "tier2_rates": {},
+        "tier2_counts": {}
+      },
+      "fuzz": {
+        "n": 20,
+        "pass_rate": null,
+        "pass_cols": [
+          "schema_ok",
+          "constraints_ok"
+        ],
+        "averages": {
+          "bucket_score": 10.0
+        },
+        "top_failures": [],
+        "signal_rates": {},
+        "signal_counts": {},
+        "tier2_rates": {},
+        "tier2_counts": {}
+      },
+      "toolcall": {
+        "n": 2,
+        "pass_rate": 1.0,
+        "pass_cols": [
+          "stage1_tool_parse_ok",
+          "stage1_tool_schema_ok"
+        ],
+        "averages": {
+          "bucket_score": 11.0
+        },
+        "top_failures": [],
+        "signal_rates": {
+          "stage1_tool_parse_ok": 1.0,
+          "stage1_tool_schema_ok": 1.0
+        },
+        "signal_counts": {
+          "stage1_tool_parse_ok": {
+            "present": 2,
+            "ok": 2,
+            "fail": 0
+          },
+          "stage1_tool_schema_ok": {
+            "present": 2,
+            "ok": 2,
+            "fail": 0
+          }
+        },
+        "tier2_rates": {},
+        "tier2_counts": {}
+      },
+      "mcq": {
+        "n": 5,
+        "pass_rate": null,
+        "pass_cols": [
+          "choice_ok"
+        ],
+        "averages": {
+          "bucket_score": 1.0
+        },
+        "top_failures": [],
+        "signal_rates": {},
+        "signal_counts": {},
+        "tier2_rates": {},
+        "tier2_counts": {}
+      }
+    }
+  },
+  "generated_from": {
+    "rows": 42
+  },
+  "version_tag": "v7.21",
+  "run_id": "20260210_215029"
+}
--- a/run_manifest.json
+++ b/run_manifest.json
@@ -0,0 +1,32 @@
+{
+  "run_id": "20260210_215029",
+  "timestamp": "20260210_215029",
+  "model_id": "Qwen/Qwen2.5-14B-Instruct-1M",
+  "hf_model_dir": "/mnt/h/Libraries/Mistralai/GGUF_Models/Qwen_Qwen2.5_14B_Instruct_1M/hf_snapshot",
+  "adapter_id": "qwen2_5_14b_instruct_1m",
+  "adapter_reason": "exact model_id match",
+  "allowed_root": "/mnt",
+  "work_root": "/mnt/h/Libraries/Mistralai",
+  "catalog_root": "/mnt/e/Libraries/Mistralai",
+  "promotion_status": "work",
+  "promotion_updated_at": "2026-02-11T03:55:50.056729Z",
+  "run_dir": "/mnt/h/Libraries/Mistralai/runs/Qwen2.5_14B_Instruct_1M/Qwen2.5_14B_Instruct_1M_20260210_215029",
+  "artifacts_dir": "/mnt/h/Libraries/Mistralai/runs/Qwen2.5_14B_Instruct_1M/Qwen2.5_14B_Instruct_1M_20260210_215029/artifacts",
+  "fixtures_dir": "/mnt/h/Libraries/Mistralai/runs/Qwen2.5_14B_Instruct_1M/Qwen2.5_14B_Instruct_1M_20260210_215029/fixtures",
+  "fixtures_file": "golden_oracle_fixtures_v7_21__sha256_6d71a0b9147c.json",
+  "fixtures_sha256": "6d71a0b9147c079371b02a94f3c149eb78a6adc03dc16ff6833b964fbf4174f0",
+  "fixtures_source_file": "golden_oracle_fixtures_v7_21.json",
+  "quant_types_requested": [],
+  "skip_quant": true,
+  "seed": 42,
+  "artifacts": [],
+  "runners": {
+    "full_weight_transformers": {
+      "name": "full_weight_transformers",
+      "reused_cache": 0,
+      "cache_reuse_path": "",
+      "cache_write_enabled": 1,
+      "cache_write_path": "/mnt/h/Libraries/Mistralai/runs/Qwen2.5_14B_Instruct_1M/Qwen2.5_14B_Instruct_1M_20260210_215029/full_weight_cache.json"
+    }
+  }
+}