--- library_name: transformers pipeline_tag: text-generation base_model: Qwen/Qwen2.5-3B-Instruct datasets: - muradil211/AetherSearch_SFT tags: - aethersearch - agentic-search - search-augmented-generation - supervised-fine-tuning - qwen2 language: - en ---

# 🔭 AetherSearch SFT ### A compact search agent that learns to reason, retrieve, and answer Fine-tuned from **Qwen2.5-3B-Instruct** on **2,000 complete search trajectories**.

[🏠 Project](https://github.com/Muradil-mamat-211/AetherSearch) · [🧪 Training code](https://github.com/Muradil-mamat-211/AetherSearch/tree/main/sft) · [📚 Dataset](https://huggingface.co/datasets/muradil211/AetherSearch_SFT)

> 🔌 **Bring your own retriever.** AetherSearch SFT is a search-agent policy, > not a self-contained QA service. The host runtime must execute each > `...` request and return evidence inside > `...`. ## ✨ Highlights - 🔎 **Search-native behavior** — learns when and what to search before answering. - 🔁 **Single- and multi-search trajectories** — trained on 1,025 single-search and 975 multi-search examples. - 🧾 **Evidence-in-the-loop reasoning** — retrieved passages stay visible as context while being excluded from the training loss. - ⚡ **Compact 3B backbone** — built on Qwen2.5-3B-Instruct for accessible experimentation and deployment. - 🧪 **Reproducible release** — public trainer, launcher, data checksum, schema tests, and artifact manifest are included or linked. ## 🧠 How it works ```text Question │ ▼ reason about what is missing │ ▼ focused retrieval query ─────► Search / RAG backend ▲ │ └──── retrieved evidence ◄────┘ │ ├── repeat the search loop when more evidence is needed ▼ evidence-grounded final answer ``` The model produces the reasoning, search, and answer spans. Your runtime owns retrieval: parse a completed `` span, run the query, append the result as ``, and resume generation until the model emits ``. ## 📊 Model at a glance | Field | Value | |---|---| | 🧱 Base model | [`Qwen/Qwen2.5-3B-Instruct`](https://huggingface.co/Qwen/Qwen2.5-3B-Instruct) | | 🧬 Base revision | `aa8e72537993ba99e69dfaafa59ed015b17504d1` | | 🏗️ Architecture | Qwen2 causal language model | | 🔢 Parameters | 3,085,938,688 | | 🎛️ Weight dtype | BF16 | | 📏 Context | 32,768 positions; training sequences capped at 4,096 | | 📚 Training data | 2,000 complete trajectories | | 🔍 Search mix | 1,025 single-search + 975 multi-search trajectories | | 🎓 Training stage | One full-trajectory SFT stage | ## 🚀 Quick start ```python import torch from transformers import AutoModelForCausalLM, AutoTokenizer model_id = "muradil211/AetherSearch_SFT" tokenizer = AutoTokenizer.from_pretrained(model_id) model = AutoModelForCausalLM.from_pretrained( model_id, torch_dtype=torch.bfloat16, device_map="auto", ) model.config.use_cache = True model.eval() ``` > 💡 Loading the checkpoint is only the first step. For end-to-end use, wrap > generation in the retrieval loop shown above and preserve the XML protocol > exactly. ## 🧬 Checkpoint identity This model was trained once on the 2,000 records in the canonical `final_sft_2000.jsonl` dataset, using the same configuration as the public AetherSearch SFT-2000 training code. The release contains the final model artifacts and reproducible code, not server-local logs or optimizer state. **Dataset SHA-256** ```text fec609652d3832c7a6c0ee2861c6f946b6cf7c3d3d40fc5d9be9b75df6325dcb ``` ## 🧪 Training recipe | Setting | Value | Setting | Value | |---|---:|---|---:| | Epochs | 1 | Learning rate | `2e-6` | | Scheduler | Cosine | Global batch size | 24 | | Precision | BF16 + TF32 | Max sequence length | 4,096 | | Padding | Dynamic | Distributed training | DeepSpeed ZeRO-3 | The training configuration matches the public SFT-2000 recipe: one epoch, learning rate `2e-6`, cosine scheduling, BF16, TF32, gradient checkpointing, dynamic padding, effective global batch size 24, and DeepSpeed ZeRO-3. On the three-worker training topology, per-device batch size 1 and gradient accumulation 8 resolve to that global batch. The completed checkpoint is exported as `final_model/`. The public launcher is hardware-topology independent: it uses the devices made visible by the surrounding runtime and derives gradient accumulation to keep global batch 24 unchanged. It does not embed physical GPU IDs, node addresses, NCCL fabric settings, allocator tuning, or server-local paths. ### 🎯 Supervision contract - ⬛ System, user, and question tokens are **masked**. - ⬛ Complete `...` spans are **masked**. - ✅ Assistant ``, ``, and `` spans are **supervised**. - ✅ The final assistant `<|im_end|>` token is **supervised**. The trainer, launcher, configuration, checksum, and schema tests are published in the [AetherSearch SFT directory](https://github.com/Muradil-mamat-211/AetherSearch/tree/main/sft). ## 📦 Files and integrity The release contains two BF16 SafeTensors shards, the shard index, model and generation configuration, tokenizer assets, this model card, the project logo, and `MODEL_MANIFEST.sha256`. It intentionally excludes optimizer states, intermediate checkpoints, `training_args.bin`, evaluation bundles, and all log files. After download, verify the release from its repository directory: ```bash sha256sum -c MODEL_MANIFEST.sha256 ``` ## ⚠️ Limitations Generated searches and answers can be incorrect, unsupported, or unsafe; retrieval and answer verification remain the caller's responsibility. No evaluation result is claimed by this model card. ## 📜 Terms No additional blanket license is asserted here. Review the [Qwen2.5-3B-Instruct license](https://huggingface.co/Qwen/Qwen2.5-3B-Instruct/blob/main/LICENSE) and the [AetherSearch SFT data attribution and rights status](https://github.com/Muradil-mamat-211/AetherSearch/blob/main/sft/ATTRIBUTION.md) before redistribution or downstream use. ---

**Built for experiments in agentic search and retrieval-augmented reasoning.** 🔎✨