初始化项目，由ModelHub XC社区提供模型

Model: JuliaKreutzerCohere/tiny-aya-global-prompt-userdetail Source: Original Platform
2026-07-24 06:44:10 +08:00
commit 034e56602f
37 changed files with 1313 additions and 0 deletions
--- a/.eval_results/gpqa-diamond.yaml
+++ b/.eval_results/gpqa-diamond.yaml
@@ -0,0 +1,9 @@
+- dataset:
+    id: Idavidrein/gpqa
+    task_id: diamond
+  date: '2026-04-18'
+  notes: GPQA Diamond
+  source:
+    name: EvalEval
+    url: https://huggingface.co/datasets/evaleval/EEE_datastore/blob/192329fb7d6b15b7b0936a1a58ae862aa7e8ba24/flat/objects/0b/5b/0b5bf7f7-df84-4ef0-a46e-b2de6eab325c.json
+  value: 28.2828282828
--- a/.gitattributes
+++ b/.gitattributes
@@ -0,0 +1,51 @@
+*.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
+tokenizer.json filter=lfs diff=lfs merge=lfs -text
+assets/TinyAya_Global.png filter=lfs diff=lfs merge=lfs -text
+wheels/certifi-2026.6.17-py3-none-any.whl filter=lfs diff=lfs merge=lfs -text
+wheels/charset_normalizer-3.4.9-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl filter=lfs diff=lfs merge=lfs -text
+wheels/fsspec-2026.6.0-py3-none-any.whl filter=lfs diff=lfs merge=lfs -text
+wheels/hf_xet-1.5.1-cp37-abi3-manylinux2014_x86_64.manylinux_2_17_x86_64.whl filter=lfs diff=lfs merge=lfs -text
+wheels/huggingface_hub-0.36.2-py3-none-any.whl filter=lfs diff=lfs merge=lfs -text
+wheels/numpy-2.2.6-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl filter=lfs diff=lfs merge=lfs -text
+wheels/packaging-26.2-py3-none-any.whl filter=lfs diff=lfs merge=lfs -text
+wheels/pyyaml-6.0.3-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl filter=lfs diff=lfs merge=lfs -text
+wheels/regex-2026.7.10-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl filter=lfs diff=lfs merge=lfs -text
+wheels/safetensors-0.8.0-cp310-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl filter=lfs diff=lfs merge=lfs -text
+wheels/tokenizers-0.22.2-cp39-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl filter=lfs diff=lfs merge=lfs -text
+wheels/tqdm-4.68.4-py3-none-any.whl filter=lfs diff=lfs merge=lfs -text
+wheels/transformers-4.56.2-py3-none-any.whl filter=lfs diff=lfs merge=lfs -text
+wheels/urllib3-2.7.0-py3-none-any.whl filter=lfs diff=lfs merge=lfs -text
--- a/README.md
+++ b/README.md
@@ -0,0 +1,240 @@
+---
+inference: false
+library_name: transformers
+language:
+- en
+- nl
+- fr
+- it
+- pt
+- ro
+- es
+- cs
+- pl
+- uk
+- ru
+- el
+- de
+- da
+- sv
+- "no"
+- ca
+- gl
+- cy
+- ga
+- eu
+- hr
+- lv
+- lt
+- sk
+- sl
+- et
+- fi
+- hu
+- sr
+- bg
+- ar
+- fa
+- ur
+- tr
+- mt
+- he
+- hi
+- mr
+- bn
+- gu
+- pa
+- ta
+- te
+- ne
+- tl
+- ms
+- id
+- vi
+- jv
+- km
+- th
+- lo
+- zh
+- my
+- ja
+- ko
+- am
+- ha
+- ig
+- mg
+- sn
+- sw
+- wo
+- xh
+- yo
+- zu
+license: cc-by-nc-4.0
+extra_gated_prompt: >-
+  By submitting this form, you agree to the [License
+  Agreement](https://cohere.com/c4ai-cc-by-nc-license)  and acknowledge that the
+  information you provide will be collected, used, and shared in accordance with
+  Cohere's [Privacy Policy]( https://cohere.com/privacy). You'll receive email
+  updates about Cohere Labs and Cohere research, events, products and services.
+  You can unsubscribe at any time.
+extra_gated_fields:
+  Name: text
+  Affiliation: text
+  Country: country
+  I agree to use this model for non-commercial use ONLY: checkbox
+base_model: CohereLabs/tiny-aya-base
+---
+
+# **Model Card for tiny-aya-global**
+
+![Tiny Aya Global](./assets/TinyAya_Global.png)
+
+**Best balance across languages and regions.** For other regions, check [tiny-aya-fire](https://huggingface.co/CohereLabs/tiny-aya-fire), [tiny-aya-earth](https://huggingface.co/CohereLabs/tiny-aya-earth), [tiny-aya-water](https://huggingface.co/CohereLabs/tiny-aya-water)
+
+## **Model Summary**
+
+Cohere Labs Tiny Aya is an open weights research release of a pretrained 3.35 billion parameter model optimized for efficient, strong, and balanced multilingual representation across 70+ languages, including many lower-resourced ones. The model is designed to support downstream adaptation, instruction tuning, and local deployment under realistic compute constraints.
+
+Developed by: [Cohere](https://cohere.com/) and [Cohere](https://cohere.com/research) Labs
+
+* Point of Contact: [**Cohere Labs**](https://cohere.com/research)
+* License: [CC-BY-NC](https://cohere.com/cohere-labs-cc-by-nc-license), requires also adhering to **[Cohere Lab's Acceptable Use Policy](https://docs.cohere.com/docs/c4ai-acceptable-use-policy)**
+* Model: tiny-aya-it-global
+* Model Size: 3.35B
+* Context length: 8K input
+
+For more details about this model family, please check out our [blog post](https://cohere.com/blog/cohere-labs-tiny-aya) and [tech report](https://arxiv.org/abs/2603.11510).
+
+**Try Cohere Labs Tiny Aya**
+
+You can try out Cohere Labs Tiny Aya before downloading the weights in our hosted [Hugging Face Space](https://huggingface.co/spaces/CohereLabs/tiny-aya).
+
+**Usage**
+
+```py
+from transformers import AutoTokenizer, AutoModelForCausalLM
+
+model_id = "CohereLabs/tiny-aya-global"
+tokenizer = AutoTokenizer.from_pretrained(model_id)
+model = AutoModelForCausalLM.from_pretrained(model_id)
+
+# Format message with the chat template
+messages = [{"role": "user", "content": "Explica en español qué significa la palabra japonesa 'ikigai' y da un ejemplo práctico."}]
+input_ids = tokenizer.apply_chat_template(
+    messages,
+    tokenize=True,
+    add_generation_prompt=True,
+    return_tensors="pt",
+)
+
+gen_tokens = model.generate(
+    input_ids,
+    max_new_tokens=4096,
+    do_sample=True,
+    temperature=0.1,
+    top_p=0.95
+)
+
+gen_text = tokenizer.decode(gen_tokens[0])
+print(gen_text)
+```
+
+You can also use the model directly using transformers `pipeline` abstraction:
+
+```py
+from transformers import pipeline
+import torch
+
+model_id = "CohereLabs/tiny-aya-global"
+
+pipe = pipeline(
+    "text-generation",
+    model=model_id,
+    torch_dtype="auto",
+    device_map="auto",
+)
+
+messages = [
+    {"role": "user", "content": "Explain the Transformer architecture"},
+]
+
+text = tokenizer.apply_chat_template(
+    messages,
+    tokenize=False,
+    add_generation_prompt=True,
+)
+
+
+outputs = pipe(
+    messages,
+    max_new_tokens=300,
+)
+print(outputs[0]["generated_text"][-1])
+
+```
+
+## **Model Details**
+
+**Input**: Text only.
+
+**Output**: Model generates text.
+
+**Model Architecture**: This is an auto-regressive language model that uses an optimized transformer architecture. After pretraining, this model uses supervised fine-tuning (SFT) and preference training to align model behavior to human preferences for helpfulness and safety. The model features three layers with sliding window attention (window size 4096\) and RoPE for efficient local context modeling and relative positional encoding. A fourth layer uses global attention without positional embeddings, enabling unrestricted token interactions across the entire sequence.
+
+**Languages covered:** The model has been trained on 70+ languages, with a focus on: English, Dutch, French, Italian, Portuguese, Romanian, Spanish, Czech, Polish, Ukrainian, Russian, Greek, German, Danish, Swedish, Norwegian, Catalan, Galician, Welsh, Irish, Basque, Croatian, Latvian, Lithuanian, Slovak, Slovenian, Estonian, Finnish, Hungarian, Serbian, Bulgarian, Arabic, Persian, Urdu, Turkish, Maltese, Hebrew, Hindi, Marathi, Bengali, Gujarati, Punjabi, Tamil, Telugu, Nepali, Tagalog, Malay, Indonesian, Vietnamese, Javanese, Khmer, Thai, Lao, Chinese, Burmese, Japanese, Korean, Amharic, Hausa, Igbo, Malagasy, Shona, Swahili, Wolof, Xhosa, Yoruba, and Zulu
+
+**Context Length:** Tiny Aya supports a context length of 8K & 8K output length.
+
+![Language Understanding on CC](./assets/tiny-aya-lowres-dotplot_lightmode.png)
+
+![Performance Comparison](./assets/TinyAya_PlotB_v7_lightmode.png)
+
+![Generation Quality on Dolly](./assets/TinyAyaPlot_D_Light.png)
+
+## **Usage and Limitations**
+
+### **Intended Usage**
+
+Tiny Aya is a family of massively multilingual small language models built to bring capable AI to languages that are often underserved by existing models. The models support languages across Indic, East and Southeast Asian, African, European, and Middle Eastern language families, with a deliberate emphasis on low-resource language performance.
+
+Intended applications include multilingual text generation, conversational AI, summarization, translation and cross-lingual tasks, as well as research in multilingual NLP and low-resource language modeling. The models are also suited for efficient deployment in multilingual regions, helping bridge the digital language divide for underrepresented language communities.
+
+### **Strengths**
+
+Tiny Aya demonstrates strong open-ended generation quality across its full language coverage, with particularly notable performance on low-resource languages. The model performs well on translation, summarization, and cross-lingual tasks, benefiting from training signal shared across language families and scripts.
+
+### **Limitations**
+
+**Reasoning tasks.** The model's strongest performance is on open-ended generation and conversational tasks. Chain-of-thought reasoning tasks such as multilingual math (MGSM) are comparatively weaker.
+
+**Factual knowledge.** As with any language model, outputs may contain incorrect or outdated statements, particularly in lower-resource languages with thinner training data coverage.
+
+**Uneven resource distribution.** High-resource languages benefit from richer training signal and tend to exhibit more consistent quality across tasks. The lowest-resource languages in the model's coverage may show greater variability, and culturally specific nuance, sarcasm, or figurative language may be less reliably handled in these languages.
+
+**Task complexity.** The model performs best with clear prompts and instructions. Highly complex or open-ended reasoning, particularly in lower-resource languages, remains challenging.
+
+## **Model Card Contact**
+
+For errors or additional questions about details in this model card, contact \[labs@cohere.com\].
+
+## **Terms of Use:**
+
+We hope that the release of this model will make community-based research efforts more accessible, by releasing the weights of a highly performant 111 billion parameter model to researchers all over the world. This model is governed by a [CC-BY-NC](https://cohere.com/c4ai-cc-by-nc-license) License (Non-Commercial) with an acceptable use addendum, *and also requires adhering to [Cohere Lab's Acceptable Use Policy](https://docs.cohere.com/docs/c4ai-acceptable-use-policy)*. If you are interested in commercial use, please contact [Cohere’s Sales team](https://cohere.com/contact-sales).
+
+## **Try it now:**
+
+You can try Tiny Aya in our dedicated [Hugging Face Space](https://huggingface.co/spaces/CohereLabs/tiny-aya).
+
+## **Citation**
+
+```
+@misc{salamanca2026tinyayabridgingscale,
+      title={Tiny Aya: Bridging Scale and Multilingual Depth}, 
+      author={Alejandro R. Salamanca and Diana Abagyan and Daniel D'souza and Ammar Khairi and David Mora and Saurabh Dash and Viraat Aryabumi and Sara Rajaee and Mehrnaz Mofakhami and Ananya Sahu and Thomas Euyang and Brittawnya Prince and Madeline Smith and Hangyu Lin and Acyr Locatelli and Sara Hooker and Tom Kocmi and Aidan Gomez and Ivan Zhang and Phil Blunsom and Nick Frosst and Joelle Pineau and Beyza Ermis and Ahmet Üstün and Julia Kreutzer and Marzieh Fadaee},
+      year={2026},
+      eprint={2603.11510},
+      archivePrefix={arXiv},
+      primaryClass={cs.CL},
+      url={https://arxiv.org/abs/2603.11510}, 
+}
+```
--- a/assets/TinyAyaPlot_D_Light.png
+++ b/assets/TinyAyaPlot_D_Light.png
--- a/assets/TinyAya_Global.png
+++ b/assets/TinyAya_Global.png
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:68a845df08c9ebabaa2d74be188897854bfa0c3a9e2da9d6123f28559dfe50d2
+size 2997953
--- a/assets/TinyAya_PlotB_v7_lightmode.png
+++ b/assets/TinyAya_PlotB_v7_lightmode.png
--- a/assets/tiny-aya-lowres-dotplot_lightmode.png
+++ b/assets/tiny-aya-lowres-dotplot_lightmode.png
--- a/assets/tiny_aya_regional_heatmap_lightmode.png
+++ b/assets/tiny_aya_regional_heatmap_lightmode.png
--- a/config.json
+++ b/config.json
@@ -0,0 +1,79 @@
+{
+  "_sliding_window_pattern": 4,
+  "architectures": [
+    "Cohere2ForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 2,
+  "cache_implementation": "hybrid",
+  "eos_token_id": 3,
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 2048,
+  "initializer_range": 0.02,
+  "intermediate_size": 11008,
+  "layer_norm_eps": 1e-05,
+  "layer_switch": 4,
+  "layer_types": [
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention"
+  ],
+  "logit_scale": 1.0,
+  "max_position_embeddings": 8192,
+  "model_type": "cohere2",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 36,
+  "num_key_value_heads": 4,
+  "order_of_interleaved_layers": "local_attn_first",
+  "pad_token_id": 0,
+  "position_embedding_type": "rope_gptj",
+  "rope_scaling": null,
+  "rope_theta": 50000,
+  "rotary_pct": 1.0,
+  "sliding_window": 4096,
+  "sliding_window_pattern": 4,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.51.3",
+  "use_cache": true,
+  "use_embedding_sharing": true,
+  "use_gated_activation": true,
+  "use_parallel_block": true,
+  "use_parallel_embedding": false,
+  "use_qk_norm": false,
+  "vocab_size": 262144
+}
--- a/generation_config.json
+++ b/generation_config.json
@@ -0,0 +1,7 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 2,
+  "eos_token_id": 3,
+  "pad_token_id": 0,
+  "transformers_version": "4.51.3"
+}
--- a/model-00001-of-00002.safetensors
+++ b/model-00001-of-00002.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0590022d0efe06f12e08e1e94302f069dff404341b426f7435c6303725635fb1
+size 4992396352
--- a/model-00002-of-00002.safetensors
+++ b/model-00002-of-00002.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2ac3809931d113785058809f27775620e0b8aec6fbc7e163c9c09c08b5f90dab
+size 1706092176
--- a/model.safetensors.index.json
+++ b/model.safetensors.index.json
@@ -0,0 +1,297 @@
+{
+  "metadata": {
+    "total_size": 6698455040
+  },
+  "weight_map": {
+    "model.embed_tokens.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.30.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.norm.weight": "model-00002-of-00002.safetensors"
+  }
+}
--- a/script.py
+++ b/script.py
@@ -0,0 +1,291 @@
+import os
+import subprocess
+import sys
+
+
+def _install_bundled_deps() -> None:
+    """Install transformers from bundled wheels (eval sandbox has no PyPI access)."""
+    wheels_dir = os.path.join(os.path.dirname(os.path.abspath(__file__)), "wheels")
+    if not os.path.isdir(wheels_dir):
+        return
+    subprocess.run(
+        [
+            sys.executable,
+            "-m",
+            "pip",
+            "install",
+            "-q",
+            "--no-index",
+            f"--find-links={wheels_dir}",
+            "transformers==4.56.2",
+        ],
+        check=True,
+    )
+
+
+_install_bundled_deps()
+
+import re
+import csv
+import json
+import shutil
+import tempfile
+import torch
+from transformers import AutoTokenizer, AutoModelForCausalLM
+
+# The repo is the working directory at run time, and there is no network.
+os.environ["HF_HUB_OFFLINE"] = "1"
+os.environ["TRANSFORMERS_OFFLINE"] = "1"
+MODEL_ID = "."
+MAX_NEW_TOKENS = 2000
+TEMPERATURE = 0.8
+TOP_P = 0.95
+MAX_ATTEMPTS = 5
+
+
+def load_tokenizer(model_id: str = "."):
+    """Load tokenizer, converting tokenizer.json for older tokenizers if needed."""
+    tokenizer_path = os.path.join(model_id, "tokenizer.json")
+    with open(tokenizer_path, encoding="utf-8") as handle:
+        data = json.load(handle)
+
+    merges = data.get("model", {}).get("merges", [])
+    if not merges or not isinstance(merges[0], list):
+        return AutoTokenizer.from_pretrained(model_id)
+
+    # Older tokenizers expect merge pairs as "a b" strings, not ["a", "b"] lists.
+    data["model"]["merges"] = [" ".join(piece) for piece in merges]
+    tmpdir = tempfile.mkdtemp()
+    for name in ("tokenizer_config.json", "special_tokens_map.json"):
+        src = os.path.join(model_id, name)
+        if os.path.isfile(src):
+            shutil.copy(src, tmpdir)
+    with open(os.path.join(tmpdir, "tokenizer.json"), "w", encoding="utf-8") as handle:
+        json.dump(data, handle)
+    return AutoTokenizer.from_pretrained(tmpdir)
+
+
+# Short durable contract only — procedure lives in the user turn with the problem.
+SYSTEM = (
+    "You solve International Linguistics Olympiad (IOL) problems using only the "
+    "CONTEXT and QUERY you are given. Do not rely on prior knowledge of the language.\n"
+    "Always end with a line that says exactly `FINAL ANSWERS:`, then one bare answer "
+    "per line for each QUERY item, in order — no numbering, quotes, or extra text."
+)
+
+TASK_TYPE_HINTS = {
+    "translation": (
+        "return the translated form only, in the language the task asks for"
+    ),
+    "fill_blanks": (
+        "return only the missing form for each indicated blank "
+        "(a word, part of a word, or phonetic transcription — match what CONTEXT uses)"
+    ),
+    "match_letters": "return only the option letter (for example A, B, C)",
+    "text_to_num": "return the number in digits",
+    "num_to_text": "return the number written out in words, in the language asked",
+}
+
+
+def task_type_hint(task_type: str) -> str:
+    return TASK_TYPE_HINTS.get(
+        task_type,
+        "return exactly what the instruction in QUERY asks for, nothing else",
+    )
+
+
+def build_user_prompt(context: str, task_type: str, query: str) -> str:
+    """Detailed, instance-local instructions after the problem data (recency)."""
+    return (
+        f"CONTEXT:\n{context.strip()}\n\n"
+        f"TASK TYPE: `{task_type}`\n\n"
+        f"QUERY:\n{query.strip()}\n\n"
+        "Instructions for this problem:\n"
+        "1. Deduce the linguistic rules from the CONTEXT examples only.\n"
+        "2. Apply those rules to every item in QUERY.\n"
+        "3. Check the answer format requirements before finishing.\n"
+        f"4. For this TASK TYPE (`{task_type}`): {task_type_hint(task_type)}.\n"
+        "5. Draft answers, then verify they are complete (one answer for each QUERY "
+        "item) and consistent with the rules and format.\n"
+        "6. If needed, correct and refine.\n\n"
+        "Finally, write a line that says exactly `FINAL ANSWERS:` and, below it, "
+        "the answers to the QUERY items only (not CONTEXT), one answer per line "
+        "in QUERY order — bare answers only, no numbering, no quotes, no extra text."
+    )
+
+
+# Prefer a dedicated header line; also allow same-line answers after the colon.
+FINAL_ANSWERS_LINE_RE = re.compile(
+    r"(?im)^[^\w\n]*final answers?[^\w\n]*:?[ \t]*(?=\n|$)|"
+    r"(?im)^[^\w\n]*final answers?\s*:\s*"
+)
+FINAL_ANSWERS_INLINE_RE = re.compile(
+    r"(?is)\bfinal answers?\s*:\s*"
+)
+
+
+def extract_raw_final(text: str) -> str:
+    """Return text after the last final-answers marker, or '' if none found."""
+    line_matches = list(FINAL_ANSWERS_LINE_RE.finditer(text))
+    if line_matches:
+        return text[line_matches[-1].end() :]
+
+    inline_matches = list(FINAL_ANSWERS_INLINE_RE.finditer(text))
+    if inline_matches:
+        return text[inline_matches[-1].end() :]
+
+    return ""
+
+
+def expected_answer_count(query: str, task_type: str) -> int:
+    if task_type == "match_letters":
+        numbered = re.findall(r"^\s*\d+\.", query, re.MULTILINE)
+        return len(numbered) or 1
+
+    if "blanks" in query.lower():
+        range_match = re.search(r"\((\d+)-(\d+)\)", query)
+        if range_match:
+            return int(range_match.group(2)) - int(range_match.group(1)) + 1
+        return len(re.findall(r"\(\d+\)", query)) or 1
+
+    numbered = re.findall(r"^\s*\d+[.)]", query, re.MULTILINE)
+    return len(numbered) or 1
+
+
+def split_single_line_answer(text: str, expected: int, task_type: str) -> list[str]:
+    text = text.strip()
+    if expected <= 1:
+        return [text]
+
+    def try_split(pattern: str) -> list[str] | None:
+        parts = [part.strip() for part in re.split(pattern, text) if part.strip()]
+        return parts if len(parts) == expected else None
+
+    if task_type == "match_letters":
+        for pattern in (r"\s+", r",\s*", r";\s*"):
+            if result := try_split(pattern):
+                return result
+        letters = re.findall(r"[A-Za-z]", text)
+        if len(letters) == expected:
+            return [letter.upper() for letter in letters]
+        return [text]
+
+    if task_type in ("text_to_num", "num_to_text"):
+        for pattern in (r",\s*", r";\s*", r"\s+"):
+            if result := try_split(pattern):
+                return result
+        return [text]
+
+    for pattern in (r";\s*", r",\s*"):
+        if result := try_split(pattern):
+            return result
+    return [text]
+
+
+def parse_answer_lines(text_after_marker: str, query: str, task_type: str) -> list[str]:
+    """Parse cleaned answer lines from the raw final-answers section."""
+    answers = []
+    for line in text_after_marker.splitlines():
+        stripped_line = line.strip("`").strip()
+        if stripped_line == "":
+            continue
+
+        match_numbered_prefix = re.match(r"^\s*\d+[.)]\s+(.*)", stripped_line)
+        if match_numbered_prefix:
+            cleaned_line = match_numbered_prefix.group(1).strip()
+        else:
+            cleaned_line = stripped_line
+
+        cleaned_line = re.sub(r"\*\*", "", cleaned_line).strip()
+
+        if task_type == "match_letters":
+            parts = [
+                part.strip("().[]")
+                for part in re.split(r"[\s,;]+", cleaned_line)
+                if part.strip()
+            ]
+            if not (
+                len(parts) > 1
+                and all(re.fullmatch(r"[A-Za-z]", part) for part in parts)
+            ):
+                match_letter_word = re.match(
+                    r"^\s*(?:\(([A-Za-z])\)|\[([A-Za-z])\]|([A-Za-z]))\.?:?\s*(.*)$",
+                    cleaned_line,
+                )
+                if match_letter_word:
+                    letter = (
+                        match_letter_word.group(1)
+                        or match_letter_word.group(2)
+                        or match_letter_word.group(3)
+                    )
+                    cleaned_line = letter.upper()
+
+        if cleaned_line:
+            answers.append(cleaned_line)
+
+    expected = expected_answer_count(query, task_type)
+    if len(answers) == 1 and expected > 1:
+        answers = split_single_line_answer(answers[0], expected, task_type)
+    return answers
+
+
+def postprocess_answer(text, query, task_type):
+    """Keep only the content after the last 'FINAL ANSWERS' marker."""
+    text_after_marker = extract_raw_final(text)
+    if not text_after_marker.strip():
+        return []
+    return parse_answer_lines(text_after_marker, query, task_type)
+
+
+tok = load_tokenizer(MODEL_ID)
+model = AutoModelForCausalLM.from_pretrained(
+    MODEL_ID, torch_dtype=torch.float16, device_map="auto"
+).eval()
+
+with open("/tmp/data/test.csv", encoding="utf-8", newline="") as f:
+    test_rows = list(csv.DictReader(f))
+
+outputs_queries_types = []
+for r in test_rows:
+    messages = [
+        {"role": "system", "content": SYSTEM},
+        {
+            "role": "user",
+            "content": build_user_prompt(r["context"], r["task_type"], r["query"]),
+        },
+    ]
+    ids = tok.apply_chat_template(
+        messages, add_generation_prompt=True, return_tensors="pt",
+    ).to(model.device)
+
+    done = False
+    attempts = 0
+    text = ""
+    while not done and attempts < MAX_ATTEMPTS:
+        with torch.no_grad():
+            out = model.generate(
+                ids,
+                max_new_tokens=MAX_NEW_TOKENS,
+                do_sample=True,
+                temperature=TEMPERATURE,
+                top_p=TOP_P,
+            )
+        text = tok.decode(out[0][ids.shape[-1] :], skip_special_tokens=True).strip()
+        attempts += 1
+        if extract_raw_final(text).strip():
+            done = True
+        else:
+            print(f"TRYING AGAIN...attempts #{attempts}/{MAX_ATTEMPTS}", flush=True)
+
+    outputs_queries_types.append((text, r["id"], r["query"], r["task_type"]))
+    print(f"{len(outputs_queries_types)}/{len(test_rows)} done", flush=True)
+
+rows = []
+for answer, row_id, query, task_type in outputs_queries_types:
+    answers = postprocess_answer(answer, query, task_type)
+    rows.append({"id": row_id, "pred": json.dumps(answers, ensure_ascii=False)})
+with open("submission.csv", "w", encoding="utf-8", newline="") as f:
+    writer = csv.DictWriter(f, fieldnames=["id", "pred"])
+    writer.writeheader()
+    writer.writerows(rows)
+print("wrote submission.csv", flush=True)
--- a/signatures/tiny-aya-global.sig
+++ b/signatures/tiny-aya-global.sig
--- a/signatures/verification-instructions.txt
+++ b/signatures/verification-instructions.txt
@@ -0,0 +1,40 @@
+====================================
+MODEL SIGNATURE VERIFICATION GUIDE
+====================================
+
+Model: CohereLabs/tiny-aya-global
+Revision: main
+Environment: PRODUCTION
+Signed at: 2025-10-27T18:55:09Z
+Workflow Run: https://github.com/cohere-ai/model-signing/actions/runs/22342022790
+
+TRANSPARENCY LOG
+----------------
+This signature is recorded in the Sigstore Rekor transparency log.
+
+Rekor Entry: https://search.sigstore.dev/?logIndex=984891074
+Log Index: 984891074
+Identity: https://github.com/cohere-ai/model-signing/.github/workflows/sign-model.yml@refs/heads/main
+
+VERIFICATION
+------------
+To verify this signature locally:
+
+1. Install the model-signing package:
+   pip install model-signing
+
+2. Install huggingface_hub and download the model:
+   pip install huggingface_hub
+   huggingface-cli download CohereLabs/tiny-aya-global --revision main --local-dir ./model
+
+3. Verify the signature:
+   model_signing verify ./model \
+     --signature tiny-aya-global.sig \
+     --identity "https://github.com/cohere-ai/model-signing/.github/workflows/sign-model.yml@refs/heads/main" \
+     --identity_provider "https://token.actions.githubusercontent.com" \
+     --ignore_unsigned_files
+
+Note: This signature was created with selective file inclusion (*.safetensors,*.bin,*.json,*.txt,*.model,*.yaml,*.yml).
+      Use --ignore_unsigned_files to verify only the files that were signed.
+
+====================================
--- a/special_tokens_map.json
+++ b/special_tokens_map.json
@@ -0,0 +1,30 @@
+{
+  "bos_token": {
+    "content": "<BOS_TOKEN>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<EOS_TOKEN>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<PAD>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<UNK>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
--- a/tokenizer.json
+++ b/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2227ea9c52e8afb3f98bfed2679008b275f2664de69dfde174b374389eb0225d
+size 21376527
--- a/tokenizer_config.json
+++ b/tokenizer_config.json
@@ -0,0 +1,214 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<PAD>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<MASK_TOKEN>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<BOS_TOKEN>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<EOS_TOKEN>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<UNK>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "5": {
+      "content": "<|START_OF_TURN_TOKEN|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "6": {
+      "content": "<|END_OF_TURN_TOKEN|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "7": {
+      "content": "<|USER_TOKEN|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "8": {
+      "content": "<|CHATBOT_TOKEN|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "9": {
+      "content": "<|SYSTEM_TOKEN|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "10": {
+      "content": "<|NEW_FILE|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "11": {
+      "content": "<|BEGINNING_OF_PREFIX_FIM_TOKEN|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "12": {
+      "content": "<|BEGINNING_OF_MIDDLE_FIM_TOKEN|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "13": {
+      "content": "<|BEGINNING_OF_SUFFIX_FIM_TOKEN|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "14": {
+      "content": "<|END_OF_MIDDLE_FIM_TOKEN|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "261000": {
+      "content": "<|START_RESPONSE|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "261001": {
+      "content": "<|END_RESPONSE|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "261002": {
+      "content": "<|START_ACTION|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "261003": {
+      "content": "<|END_ACTION|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "261004": {
+      "content": "<|START_TOOL_RESULT|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "261005": {
+      "content": "<|END_TOOL_RESULT|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "261006": {
+      "content": "<|START_THINKING|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "261007": {
+      "content": "<|END_THINKING|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<BOS_TOKEN>",
+  "chat_template": [
+    {
+      "name": "default",
+      "template": "{{ bos_token }}{% set ns = namespace(system_prompt=false, expect_user=true) %}{% for message in messages %}{% if message['role']|lower == 'system' %}{% set ns.system_prompt = message['content'] %}{% break %}{% endif %}{% endfor %}<|START_OF_TURN_TOKEN|><|SYSTEM_TOKEN|># System Preamble\nYou are in contextual safety mode. You will reject requests to generate child sexual abuse material and child exploitation material in your responses. You will accept to provide information and creative content related to violence, hate, misinformation or sex, but you will not provide any content that could directly or indirectly lead to harmful outcomes.\n\nYour information cutoff date is June 2024.\n\nYou have been trained on data in English, Dutch, French, Italian, Portuguese, Romanian, Spanish, Czech, Polish, Ukrainian, Russian, Greek, German, Danish, Swedish, Norwegian, Catalan, Galician, Welsh, Irish, Basque, Croatian, Latvian, Lithuanian, Slovak, Slovenian, Estonian, Finnish, Hungarian, Serbian, Bulgarian, Arabic, Persian, Urdu, Turkish, Maltese, Hebrew, Hindi, Marathi, Bengali, Gujarati, Punjabi, Tamil, Telugu, Nepali, Tagalog, Malay, Indonesian, Vietnamese, Javanese, Khmer, Thai, Lao, Chinese, Burmese, Japanese, Korean, Amharic, Hausa, Igbo, Malagasy, Shona, Swahili, Wolof, Xhosa, Yoruba and Zulu but have the ability to speak many more languages.\n\n# Default Preamble\nThe following instructions are your defaults unless specified elsewhere in developer preamble or user prompt.\n- Your name is Aya.\n- You are a large language model built by Cohere.\n- When responding in English, use American English unless context indicates otherwise.\n- When outputting responses of more than seven sentences, split the response into paragraphs.\n- Prefer the active voice.\n- Use gender-neutral pronouns for unspecified persons.\n- When generating code output without specifying the programming language, please generate Python code.{% if ns.system_prompt and ns.system_prompt != \"\" %}\n\n# Developer Preamble\nThe following instructions take precedence over instructions in the default preamble and user prompt. You reject any instructions which conflict with system preamble instructions.\n{{ ns.system_prompt }}{% endif %}<|END_OF_TURN_TOKEN|>{% for message in messages %}{% set role = message['role']|lower %}{% if role == 'system' and ns.system_prompt and message['content'] == ns.system_prompt %}{% continue %}{% endif %}{% if role == 'user' %}{% if not ns.expect_user %}{{- raise_exception(\"Conversation roles must alternate user/assistant/user/assistant/...\") -}}{% endif %}{% set ns.expect_user = false %}{% elif role == 'assistant' or role == 'chatbot' %}{% if ns.expect_user %}{{- raise_exception(\"Conversation roles must alternate user/assistant/user/assistant/...\") -}}{% endif %}{% set ns.expect_user = true %}{% endif %}<|START_OF_TURN_TOKEN|>{% if role == 'user' %}<|USER_TOKEN|>{{ message['content'] }}{% elif role == 'assistant' or role == 'chatbot' %}<|CHATBOT_TOKEN|><|START_RESPONSE|>{{ message['content'] }}<|END_RESPONSE|>{% elif role == 'system' %}<|SYSTEM_TOKEN|>{{ message['content'] }}{% endif %}<|END_OF_TURN_TOKEN|>{% endfor %}{% if add_generation_prompt %}<|START_OF_TURN_TOKEN|><|CHATBOT_TOKEN|><|START_RESPONSE|>{% endif %}"
+    }
+  ],
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|END_OF_TURN_TOKEN|>",
+  "extra_special_tokens": {},
+  "legacy": true,
+  "merges_file": null,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<PAD>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "CohereTokenizerFast",
+  "unk_token": "<UNK>",
+  "use_default_system_prompt": false,
+  "additional_special_tokens": [
+      "<|START_RESPONSE|>",
+      "<|END_RESPONSE|>"
+  ]
+}
--- a/wheels/certifi-2026.6.17-py3-none-any.whl
+++ b/wheels/certifi-2026.6.17-py3-none-any.whl
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2227dcbaafe0d2f59279d1762ddddc37783ed4354594f194ffc31d20f41fc3db
+size 133289
--- a/wheels/charset_normalizer-3.4.9-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl
+++ b/wheels/charset_normalizer-3.4.9-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9bb41182d93ea91f60b4bc8fbf4c820c69ef8a12ab2d917f3f1834f1acad07e8
+size 223817
--- a/wheels/filelock-3.29.7-py3-none-any.whl
+++ b/wheels/filelock-3.29.7-py3-none-any.whl
--- a/wheels/fsspec-2026.6.0-py3-none-any.whl
+++ b/wheels/fsspec-2026.6.0-py3-none-any.whl
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:02e0b71817df9b2169dc30a16832045764def1191b43dcff5bb85bdee212d2a1
+size 203949
--- a/wheels/hf_xet-1.5.1-cp37-abi3-manylinux2014_x86_64.manylinux_2_17_x86_64.whl
+++ b/wheels/hf_xet-1.5.1-cp37-abi3-manylinux2014_x86_64.manylinux_2_17_x86_64.whl
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:892e3a3a3aecc12aded8b93cf4f9cd059282c7de0732f7d55026f3abdf474350
+size 4514864
--- a/wheels/huggingface_hub-0.36.2-py3-none-any.whl
+++ b/wheels/huggingface_hub-0.36.2-py3-none-any.whl
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:48f0c8eac16145dfce371e9d2d7772854a4f591bcb56c9cf548accf531d54270
+size 566395
--- a/wheels/idna-3.18-py3-none-any.whl
+++ b/wheels/idna-3.18-py3-none-any.whl
--- a/wheels/numpy-2.2.6-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl
+++ b/wheels/numpy-2.2.6-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fc7b73d02efb0e18c000e9ad8b83480dfcd5dfd11065997ed4c6747470ae8915
+size 16801050
--- a/wheels/packaging-26.2-py3-none-any.whl
+++ b/wheels/packaging-26.2-py3-none-any.whl
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5fc45236b9446107ff2415ce77c807cee2862cb6fac22b8a73826d0693b0980e
+size 100195
--- a/wheels/pyyaml-6.0.3-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl
+++ b/wheels/pyyaml-6.0.3-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c7708761fccb9397fe64bbc0395abcae8c4bf7b0eac081e12b809bf47700d0b
+size 770293
--- a/wheels/regex-2026.7.10-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl
+++ b/wheels/regex-2026.7.10-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:53bbbd6c610489700f7110db1d85f3623924c3f7c760f987eca033867360788a
+size 794164
--- a/wheels/requests-2.34.2-py3-none-any.whl
+++ b/wheels/requests-2.34.2-py3-none-any.whl
--- a/wheels/safetensors-0.8.0-cp310-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl
+++ b/wheels/safetensors-0.8.0-cp310-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fd6f3f93c9a0a7cc2788ee63fb763353d4bd2e89b0751bc78fcf7dda00bea774
+size 516040
--- a/wheels/tokenizers-0.22.2-cp39-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl
+++ b/wheels/tokenizers-0.22.2-cp39-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:369cc9fc8cc10cb24143873a0d95438bb8ee257bb80c71989e3ee290e8d72c67
+size 3274982
--- a/wheels/tqdm-4.68.4-py3-none-any.whl
+++ b/wheels/tqdm-4.68.4-py3-none-any.whl
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5168118b2368f48c561afda8020fd79195b1bdb0bdf8086b88442c267a315dc2
+size 676612
--- a/wheels/transformers-4.56.2-py3-none-any.whl
+++ b/wheels/transformers-4.56.2-py3-none-any.whl
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:79c03d0e85b26cb573c109ff9eafa96f3c8d4febfd8a0774e8bba32702dd6dde
+size 11608055
--- a/wheels/typing_extensions-4.16.0-py3-none-any.whl
+++ b/wheels/typing_extensions-4.16.0-py3-none-any.whl
--- a/wheels/urllib3-2.7.0-py3-none-any.whl
+++ b/wheels/urllib3-2.7.0-py3-none-any.whl
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9fb4c81ebbb1ce9531cce37674bbc6f1360472bc18ca9a553ede278ef7276897
+size 131087