Downloads · 30 days
4
4% of all-time downloads
satoyutaka/Qwen2.5-7B-AgentBench-V4-BF16
Qwen2.5-7B-AgentBench-V4-BF16 is a machine learning model from satoyutaka. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
This is the V4 Advanced model, a high-precision variant of the agent model based on the Qwen2.5-7B-Instruct architecture, specifically designed to achieve extreme accuracy and long-context understanding in the AgentBe…
Downloads · 30 days
4
4% of all-time downloads
All-time downloads
100
Public
Parameters
7.6B
15.2 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors15.2 GB · 100%
From the Hugging Face model README
This is the V4 Advanced model, a high-precision variant of the agent model based on the Qwen2.5-7B-Instruct architecture, specifically designed to achieve extreme accuracy and long-context understanding in the AgentBench-comp evaluation environment.
The reasoning capability has been maximize through strict data curation and extended context length, aiming to solve complex multi-step tasks without generation errors.
| V3-BF16 | V4-BF16 | |
|---|---|---|
| Context Length | 2048 | 4096 (Handles longer ALFWorld trajectories) |
| Dataset | Standard SFT | Iron Guard (171 strictly curated high-quality trajectories) |
| Training Method | Validation enabled | Optimized SFT (Batch 1, GradAcc 4, No validation overhead) |
| Target Focus | General Agent | Aggregations & Complex Planning (SQL SUM/COUNT, long navigation) |
0.192 through 1500 intensive iterations on Mac M4.V4 was born from the need to eliminate formatting errors and improve strict adherence in long tasks:
max_seq_length to 4096.batch_size=1, grad_accumulation=4, and grad_checkpointing.# Evaluate directly in BF16 format (if within timeout limits)
cd ~/AgentBench
./update_model.sh satoyutaka/Qwen2.5-7B-AgentBench-V4-BF16
python3 -m src.assigner -c configs/assignments/dbbench-vllm.yaml
python3 -m src.assigner -c configs/assignments/alfworld-vllm.yaml
本モデルは、AgentBench-comp コンペティションの評価環境において「極限の精度」と「長文脈の理解力」を達成するために設計された、V4 アドバンス版エージェントモデルです(Qwen2.5-7B-Instructベース)。
厳格なデータ選別とコンテキスト長の拡張により推論能力を最大化し、複雑な多段階タスクを生成エラーなしで解決することを目指しています。
V4は、フォーマットエラーを完全に排除し、長期タスクでの厳密な指示順守を向上させる必要性から生まれました:
max_seq_lengthを4096に倍増しました。batch_size=1, grad_accumulation=4, grad_checkpointing を駆使し、MacのOOMエラーを克服しました。0.192)。Author: satoyutaka Competition: AgentBench-comp (V4-BF16) Base Model: Qwen2.5-7B-Instruct (LoRA fine-tuned, Iter 1500) Precision: BFloat16