---
license: apache-2.0
library_name: transformers
pipeline_tag: text-generation
base_model:
- Qwen/Qwen3-8B
datasets:
- TIGER-Lab/FIM-Midtraining-400K
- SWE-Lego/SWE-Lego-Synthetic-Data
tags:
- code
- software-engineering
- agent
---

# FIM-8B

[📄 Paper](https://arxiv.org/abs/2607.12463) · [💻 GitHub](https://github.com/TIGER-AI-Lab/FIM-Midtraining) · [🤗 Dataset](https://huggingface.co/datasets/TIGER-Lab/FIM-Midtraining-400K) · [🤗 Collection](https://huggingface.co/collections/TIGER-Lab/fim-midtraining)

**FIM-8B** is the strongest released model of *"Function-Aware Fill-in-the-Middle as Mid-Training for Coding Agent Foundation Models"*: `Qwen3-8B`, mid-trained on function-aware FIM data, then post-trained on SWE-Lego agent trajectories. The mid-training stage is the only difference from a standard SWE-Lego reproduction — worth **+3.2 points on SWE-Bench-Verified and +5.4 on SWE-Bench-Lite**. Unlike FIM-7B and FIM-14B (R2E-Gym scaffold), this model is evaluated with the SWE-Lego setup: OpenHands `CodeActAgent` for inference and the official SWE-bench harness for scoring.

## Training pipeline

- **Base model**: [`Qwen/Qwen3-8B`](https://huggingface.co/Qwen/Qwen3-8B)
- **FIM mid-training**: [`midtraining/configs/fim_midtrain.yaml`](https://github.com/TIGER-AI-Lab/FIM-Midtraining/blob/main/midtraining/configs/fim_midtrain.yaml) on [TIGER-Lab/FIM-Midtraining-400K](https://huggingface.co/datasets/TIGER-Lab/FIM-Midtraining-400K) (as-run copy: [`FIM_Midtrain_8B.yaml`](https://github.com/TIGER-AI-Lab/FIM-Midtraining/blob/main/midtraining/configs/FIM_Midtrain_8B.yaml)) → intermediate checkpoint released as [TIGER-Lab/FIM-Mid-8B](https://huggingface.co/TIGER-Lab/FIM-Mid-8B)
- **Post-training**: SFT on SWE-Lego trajectories (real + synthetic, `resolved`-filtered, 2 epochs) — [`posttraining/swe_lego/`](https://github.com/TIGER-AI-Lab/FIM-Midtraining/tree/main/posttraining/swe_lego) (as-run copy: [`FIM_Posttrain_8B.yaml`](https://github.com/TIGER-AI-Lab/FIM-Midtraining/blob/main/posttraining/swe_lego/FIM_Posttrain_8B.yaml))

## Results

Means over three evaluation seeds, identical harness for both arms (paper Table 1):

| Setting | SWE-Bench-Verified | SWE-Bench-Lite |
|---|---|---|
| Qwen3-8B + SWE-Lego (reproduced) | 31.80 | 27.30 |
| **FIM-8B (+ FIM mid-training)** | **35.00** | **32.70** |
| Δ | +3.20 | +5.40 |

## Evaluate on SWE-Bench Verified

The complete pinned walkthrough lives at [`evaluation/swebench/released_checkpoints.md`](https://github.com/TIGER-AI-Lab/FIM-Midtraining/blob/main/evaluation/swebench/released_checkpoints.md).

### 1. Serve the model with vLLM

The checkpoint ships `max_position_embeddings: 163840` and its own chat template, so no rope or template overrides are needed:

```bash
CUDA_VISIBLE_DEVICES=0 \
python -m vllm.entrypoints.openai.api_server \
  --model TIGER-Lab/FIM-8B \
  --served-model-name FIM-8B \
  --host 127.0.0.1 \
  --port 8400 \
  --tensor-parallel-size 1 \
  --max-model-len 163840 \
  --max-num-seqs 16 \
  --gpu-memory-utilization 0.9 \
  > vllm_fim8b.log 2>&1 &
```

Wait until the server is up (model load takes ~1 minute):

```bash
curl -s http://127.0.0.1:8400/v1/models
```

### 2. Run the agent on SWE-Bench Verified

Inference uses OpenHands 0.53.0 with `CodeActAgent`. Define the LLM in `config.toml`:

```toml
[llm.eval_fim]
model = "openai/FIM-8B"
base_url = "http://127.0.0.1:8400/v1"
api_key = "EMPTY"
temperature = 0.0
max_input_tokens = 147456
max_output_tokens = 16384
native_tool_calling = false
```

From the OpenHands checkout:

```bash
env USE_HINT_TEXT=false \
    INSTRUCTION_TEMPLATE_NAME=swe_default.j2 \
    ENABLE_PLAN_MODE=false \
    ADD_IN_CONTEXT_LEARNING_EXAMPLE=false \
poetry run python evaluation/benchmarks/swe_bench/run_infer.py \
  --config-file config.toml \
  --agent-cls CodeActAgent \
  --llm-config llm.eval_fim \
  --max-iterations 100 \
  --eval-num-workers 1 \
  --eval-output-dir ./eval_out \
  --dataset princeton-nlp/SWE-bench_Verified \
  --split test \
  --mode swe
```

For SWE-Bench Lite, use `--dataset princeton-nlp/SWE-bench_Lite`.

### 3. Score with the SWE-bench harness

Convert the OpenHands `output.jsonl` to a predictions file with `evaluation/benchmarks/swe_bench/scripts/eval/convert_oh_output_to_swe_json.py`, then evaluate it with the official SWE-bench harness (`python -m swebench.harness.run_evaluation`). The reported score is `resolved_instances / total_instances`.

## Citation

```bibtex
@article{wang2026fim,
  title={Function-Aware Fill-in-the-Middle as Mid-Training for Coding Agent Foundation Models},
  author={Wang, Yubo and Liang, Jiarong and Zhang, Yuxuan and Liu, Xuye and Wei, Cong and Zhang, Yuyu and Nie, Ping and Chen, Wenhu},
  journal={arXiv preprint arXiv:2607.12463},
  year={2026}
}
```