---
library_name: transformers
tags:
- small-lm
- math
- code
- reasoning
- slm
- french
- merge
license: apache-2.0
base_model:
- SamsungSAILMontreal/Qwen3-4B-Instruct-2507
- SamsungSAILMontreal/Qwen3-4B-Instruct-2507-Math
- SamsungSAILMontreal/Qwen3-4B-Instruct-2507-Code
- SamsungSAILMontreal/Qwen3-4B-Instruct-2507-Fr
language:
- fr
- en
---

# Qwen3-4B-Instruct-2507-Math-Fr

This model is obtained by merging [SamsungSAILMontreal/Qwen3-4B-Instruct-2507-Math](https://huggingface.co/SamsungSAILMontreal/Qwen3-4B-Instruct-2507-Math), 
[SamsungSAILMontreal/Qwen3-4B-Instruct-2507-Code](https://huggingface.co/SamsungSAILMontreal/Qwen3-4B-Instruct-2507-Code) and 
[SamsungSAILMontreal/Qwen3-4B-Instruct-2507-Fr](https://huggingface.co/SamsungSAILMontreal/Qwen3-4B-Instruct-2507-Fr).
The model is used in the experiments described in https://bknyaz.github.io/blog/2026/meta-merge/. 
Single A100 was used for merging and evaluation.

The following versions were used for merge/eval:

- python >= 3.10
- torch               : 2.9.0+cu128
- lm_eval             : 0.4.9.1
- vllm                : 0.11.1
- transformers        : 4.57.6
- datasets            : 3.2.0
- numpy               : 2.2.6

## Merging

Merging was done using parameter averaging implemented in [merge_qwen.py](https://github.com/SamsungSAILMontreal/nino/blob/main/merge_qwen.py).

## Evaluation

Evaluation was done with lm_eval on the test split of [gsm8k](https://huggingface.co/datasets/openai/gsm8k), 
[french_bench (avg score)](https://github.com/EleutherAI/lm-evaluation-harness/tree/main/lm_eval/tasks/french_bench), 
[gsm8k-fr](https://huggingface.co/datasets/cmh/gsm8k_fr) and humaneval (instruct):

```bash
python -m lm_eval --model vllm --model_args pretrained=${model},tensor_parallel_size=1,dtype=auto,gpu_memory_utilization=0.9,data_parallel_size=1 \
 --tasks gsm8k,french_bench,gsm8k-fr,humaneval_instruct --batch_size 1 --apply_chat_template=True --confirm_run_unsafe_code --trust_remote_code
```

To evaluate on gsm8k-fr you can use our fork https://github.com/bknyaz/lm-evaluation-harness/tree/main/lm_eval/tasks/gsm8k. 

### Results

| Model                               | gsm8k  | french | gsm8k-fr | humaneval_instruct | avg  |
|-------------------------------------|--------|--------|----------|--------------------|------|
| Qwen3-4B-Instruct-2507              | 80.4   | 43.1   | 66.0     | 90.2               | 69.9 |
| Qwen3-4B-Instruct-2507-Math         | 76.8   | 43.0   | 65.3     | 72.0               | 64.3 |
| Qwen3-4B-Instruct-2507-Fr           | 72.3   | 45.7   | 60.7     | 74.4               | 63.3 |
| Qwen3-4B-Instruct-2507-Code         | 72.5   | 45.4   | 53.0     | 76.2               | 61.8 |
| Qwen3-4B-Instruct-2507-Math-Code-Fr | 82.9   | 45.8   | 69.8     | 79.9               | 69.6 |

## License

Please refer to the license of the base models [SamsungSAILMontreal/Qwen3-4B-Instruct-2507-Math](https://huggingface.co/SamsungSAILMontreal/Qwen3-4B-Instruct-2507-Math), 
[SamsungSAILMontreal/Qwen3-4B-Instruct-2507-Code](https://huggingface.co/SamsungSAILMontreal/Qwen3-4B-Instruct-2507-Code) and 
[SamsungSAILMontreal/Qwen3-4B-Instruct-2507-Fr](https://huggingface.co/SamsungSAILMontreal/Qwen3-4B-Instruct-2507-Fr).