Qwen3-1.7B-Coder-Distilled-SFT-GGUF/README.md

---
library_name: llama.cpp
license: apache-2.0
language:
  - en
base_model: reaperdoesntknow/Qwen3-1.7B-Coder-Distilled-SFT
tags:
  - gguf
  - quantized
  - distillation
  - sft
  - reasoning
  - mathematics
  - physics
  - logic
  - logical-inference
  - stem
  - code
  - chain-of-thought
  - convergentintel
  - edge
  - knowledge-distillation
pipeline_tag: text-generation
---

# Qwen3-1.7B-Coder-Distilled-SFT — GGUF

GGUF quantizations of [reaperdoesntknow/Qwen3-1.7B-Coder-Distilled-SFT](https://huggingface.co/reaperdoesntknow/Qwen3-1.7B-Coder-Distilled-SFT) for local and edge deployment via [llama.cpp](https://github.com/ggerganov/llama.cpp) and compatible runtimes.

Coder teacher → STEM distillation → logical inference SFT → quantized. Structured reasoning in ~1.2GB.

## Available Quantizations

| File | Quant | Size | Use Case |
|---|---|---|---|
| `qwen3-1.7b-coder-distilled-sft-f16.gguf` | F16 | ~3.8 GB | Full precision reference |
| `qwen3-1.7b-coder-distilled-sft-Q8_0.gguf` | Q8_0 | ~2.1 GB | Near-lossless, desktop |
| `qwen3-1.7b-coder-distilled-sft-Q5_K_M.gguf` | Q5_K_M | ~1.4 GB | Balanced quality and size |
| `qwen3-1.7b-coder-distilled-sft-Q4_K_M.gguf` | Q4_K_M | ~1.2 GB | Mobile, edge, fastest inference |

**Recommended:** Q5_K_M for desktop, Q4_K_M for mobile/edge.

## About the Model

Two-stage build:

**Stage 1 — Coder Teacher Distillation:** Qwen3-1.7B distilled from [Qwen3-Coder-30B-A3B-Instruct](https://huggingface.co/Qwen/Qwen3-Coder-30B-A3B-Instruct) on 6,122 STEM CoT samples. Proof-weighted cross-entropy (2.5x → 1.5x on derivation tokens) + KL divergence at T=2.0. The Coder teacher transfers structured decomposition patterns — sequential logic, state tracking, compositional reasoning — through the softmax landscape.

**Stage 2 — Logical Inference SFT:** Fine-tuned on [KonstantinDob/logic_inference_dataset](https://huggingface.co/datasets/KonstantinDob/logic_inference_dataset) (~54,607 propositional logic pairs, LOGICINFERENCEe format). The model performs inference first, then concludes. Based on the [LogicInference paper](https://openreview.net/pdf?id=HAGeIS_Lcg9) by Santiago Ontañón (Google Research).

| Attribute | Value |
|---|---|
| **Base model** | [Qwen/Qwen3-1.7B](https://huggingface.co/Qwen/Qwen3-1.7B) |
| **Teacher model** | [Qwen/Qwen3-Coder-30B-A3B-Instruct](https://huggingface.co/Qwen/Qwen3-Coder-30B-A3B-Instruct) |
| **Stage 1 data** | 6,122 STEM CoT samples |
| **Stage 2 data** | ~54,607 logical inference pairs |
| **Developer** | Reaperdoesntrun / [Convergent Intelligence LLC](https://convergentintel.com): Research Division |

## Usage

### llama.cpp CLI

```bash
./llama-cli -m qwen3-1.7b-coder-distilled-sft-Q4_K_M.gguf \
  -p "### Instruction:\nConsider the premises: If it rains, the ground is wet. It is raining. What can we conclude?\n\n### Response:\n" \
  -n 512 --temp 0.0
```

### llama.cpp Python

```python
from llama_cpp import Llama

llm = Llama(model_path="qwen3-1.7b-coder-distilled-sft-Q4_K_M.gguf", n_ctx=1024)

output = llm(
    "### Instruction:\nIs the following argument valid? All dogs are animals. Some animals are pets. Therefore, all dogs are pets.\n\n### Response:\n",
    max_tokens=512,
    temperature=0.0,
)
print(output["choices"][0]["text"])
```

### Ollama

```bash
echo 'FROM ./qwen3-1.7b-coder-distilled-sft-Q4_K_M.gguf' > Modelfile
ollama create logic-reasoner -f Modelfile
ollama run logic-reasoner "If all humans are mortal and Socrates is human, what follows?"
```

### LM Studio

Download any GGUF file and load directly in [LM Studio](https://lmstudio.ai/).

## Prompt Formats

**STEM derivation (Stage 1):**

```
Solve the following problem carefully and show a rigorous derivation.

Problem:
[Your problem]

Proof:
```

**Logical inference / instruction-following (Stage 2):**

```
### Instruction:
[Your question or logical inference problem]

### Response:
```

## Limitations

1.7B model. Structured reasoning with hard capacity limits. Not a code generator despite the Coder teacher. Not a formal proof verifier. Complex multi-step inferences with many quantifiers may exceed capacity. Always verify critical outputs.

## Source Model

Full training methodology at: **[reaperdoesntknow/Qwen3-1.7B-Coder-Distilled-SFT](https://huggingface.co/reaperdoesntknow/Qwen3-1.7B-Coder-Distilled-SFT)**


## Mathematical Foundations

This is a GGUF-quantized variant. The mathematical foundations (Discrepancy Calculus, Topological Knowledge Distillation) are documented in the source model's card. The discrepancy operator $Df(x)$ and BV decomposition that inform the training pipeline are preserved through quantization — the structural boundaries detected by DISC during training are baked into the weights, not dependent on precision.

## Related Models

| Model | Description |
|---|---|
| [Qwen3-1.7B-Coder-Distilled](https://huggingface.co/reaperdoesntknow/Qwen3-1.7B-Coder-Distilled) | Stage 1 only |
| [Qwen3-1.7B-Coder-Distilled-SFT](https://huggingface.co/reaperdoesntknow/Qwen3-1.7B-Coder-Distilled-SFT) | Full precision source |
| [Qwen3-1.7B-Distilled-30B-A3B-SFT-GGUF](https://huggingface.co/reaperdoesntknow/Qwen3-1.7B-Distilled-30B-A3B-SFT-GGUF) | Instruct teacher + legal SFT GGUF |

## Citation

```bibtex
@misc{colca2026codersftgguf,
  title={Coder-Distilled Logical Inference GGUF: Structured Reasoning for Edge Deployment},
  year={2026},
  publisher={HuggingFace},
  url={https://huggingface.co/reaperdoesntknow/Qwen3-1.7B-Coder-Distilled-SFT-GGUF},
  note={Convergent Intelligence LLC: Research Division}
}
```

---

*Convergent Intelligence LLC: Research Division*
*"Where classical analysis fails to see, we begin."*

---

## Convergent Intelligence Portfolio

*Part of the [Qwen3 Coder Series](https://huggingface.co/reaperdoesntknow) by [Convergent Intelligence LLC: Research Division](https://huggingface.co/reaperdoesntknow)*


#
## Mathematical Foundations

This is a GGUF-quantized variant. The mathematical foundations (Discrepancy Calculus, Topological Knowledge Distillation) are documented in the source model's card. The discrepancy operator $Df(x)$ and BV decomposition that inform the training pipeline are preserved through quantization — the structural boundaries detected by DISC during training are baked into the weights, not dependent on precision.

## Related Models

| Model | Downloads | Format |
|-------|-----------|--------|
| [Qwen3-1.7B-Coder-Distilled-SFT](https://huggingface.co/reaperdoesntknow/Qwen3-1.7B-Coder-Distilled-SFT) | 302 | HF |

### Top Models from Our Lab

| Model | Downloads |
|-------|-----------|
| [Qwen3-1.7B-Thinking-Distil](https://huggingface.co/reaperdoesntknow/Qwen3-1.7B-Thinking-Distil) | 501 |
| [LFM2.5-1.2B-Distilled-SFT](https://huggingface.co/reaperdoesntknow/LFM2.5-1.2B-Distilled-SFT) | 342 |
| [Qwen3-0.6B-Distilled-30B-A3B-Thinking-SFT-GGUF](https://huggingface.co/reaperdoesntknow/Qwen3-0.6B-Distilled-30B-A3B-Thinking-SFT-GGUF) | 203 |
| [Qwen3-1.7B-Distilled-30B-A3B-SFT-GGUF](https://huggingface.co/reaperdoesntknow/Qwen3-1.7B-Distilled-30B-A3B-SFT-GGUF) | 175 |
| [SMOLM2Prover-GGUF](https://huggingface.co/reaperdoesntknow/SMOLM2Prover-GGUF) | 150 |

**Total Portfolio: 41 models | 2,781 total downloads**


*Last updated: 2026-03-28 12:49 UTC*

<!-- DISTILQWEN-SPOTLIGHT-START -->

## DistilQwen Collection

This model is part of the **[DistilQwen](https://huggingface.co/collections/reaperdoesntknow/distilqwen-69bf40ec669117e3f069ef1c)** proof-weighted distillation series.
Collection: **9 models** | **2,788 downloads**

### Teacher Variant Comparison

| Teacher | Student Size | Strength | Models |
|---------|-------------|----------|--------|
| Qwen3-30B-A3B (Instruct) | 1.7B | Instruction following, structured output, legal reasoning | 3 (833 DL) |
| Qwen3-30B-A3B (Thinking) | 0.6B | Extended deliberation, higher-entropy distributions, proof derivation | 3 (779 DL) |
| Qwen3-30B-A3B (Coder) | 1.7B | Structured decomposition, STEM derivation, logical inference | 2 (825 DL) **← this model** |

### Methodology

**The only BF16 collection in the portfolio.** While the broader Convergent Intelligence catalog (43 models, 12,000+ downloads) was trained on CPU at FP32 for $24 total compute, the DistilQwen series was trained on H100 at BF16 with a 30B-parameter teacher. Same methodology, premium hardware. This is what happens when you give the pipeline real compute.

All models use proof-weighted knowledge distillation: 55% cross-entropy with decaying proof weights (2.5× → 1.5×), 45% KL divergence at T=2.0. The proof weight amplifies loss on reasoning-critical tokens, forcing the student to allocate capacity to structural understanding rather than surface-level pattern matching.

Full methodology: [Structure Over Scale (DOI: 10.57967/hf/8165)](https://doi.org/10.57967/hf/8165)

### Related in this series

- [Qwen3-1.7B-Coder-Distilled-SFT](https://huggingface.co/reaperdoesntknow/Qwen3-1.7B-Coder-Distilled-SFT) (508 downloads)

<!-- DISTILQWEN-SPOTLIGHT-END -->

---
<sub>Part of the [reaperdoesntknow research portfolio](https://huggingface.co/reaperdoesntknow) — 49 models, 22,598 total downloads | Last refreshed: 2026-03-30 12:05 UTC</sub>
<!-- cix-keeper-ts:2026-04-11T16:09:15Z -->
<!-- card-refresh: 2026-03-30 -->