georvn7/hayabusa-9b

Hayabusa 9B

Hayabusa 9B is a text-only, full-weight fine-tune of Qwen/Qwen3.5-9B for software debugging and structured agent action selection.

This repository contains a full merged BF16 Transformers checkpoint, not a LoRA, QLoRA, or adapter. It can be loaded directly by Transformers-compatible runtimes or used as the source checkpoint for MLX conversion.

Current Checkpoint

hayabusa-9b-spark-pre-stochastic-restore-step3-r3-c6-online-sft-judge-v1-reboot-20261022a

Completed step3 recovery round 3, Cycle 6 final SFT, exported on 2026-10-23 UTC. Checkpoint identity: eb4adaac059c266dedabe5fedce94876077ba0d00808ffefab793c5faa79eeb1.

• Cycle 6 training: 163 DPO updates, then 280 RL updates, then 81 original teacher SFT updates.
• The published weights are the completed final SFT, not the unfinished round 4 training run.
• Full-weight BF16 safetensors: 32 shards, 427 model tensors. Stored model weights are exported without folding optimizer residuals.
• Training context ceiling: 32K tokens with left prompt truncation. Loss covers only the final assistant reasoning and content; earlier messages are masked.
• Saved architecture: Qwen3_5ForCausalLM / qwen3_5_text; recorded Transformers version 5.3.0.dev0.
• Weights and tokenizer only: no optimizer, scheduler, RNG state, private training examples or execution logs are published.

For a stable revision, use step3-r3-c6-final-sft-eb4adaac. SHA256SUMS and checkpoint_provenance.json describe the published files and completed training lineage. Successful export and training audits are integrity checks, not behavioral-quality or independent-mastery claims. This release does not claim that step3 is mastered.

Intended Use

Hayabusa is specialized for structured debugger contexts containing source, logs, traces, test outcomes, compact progress history, and constrained actions. It is not intended as a general-purpose chat model.

The checkpoint is suitable for download and conversion by MLX tooling because all model weights, tokenizer files, chat template, generation configuration, and safetensors index are present at repository root.

Loading With Transformers

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "georvn7/hayabusa-9b"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    torch_dtype=torch.bfloat16,
    device_map="auto",
)

messages = [{"role": "user", "content": "Select the next grounded debugging action."}]
text = tokenizer.apply_chat_template(
    messages,
    tokenize=False,
    add_generation_prompt=True,
    enable_thinking=True,
)
inputs = tokenizer(text, return_tensors="pt").to(model.device)
output_ids = model.generate(**inputs, max_new_tokens=512, do_sample=False)
print(tokenizer.decode(output_ids[0], skip_special_tokens=True))

Limitations

• Specialized debugging/action model, not a broad assistant benchmark release.
• Best results require the structured runtime context and action schema used during training.
• The model remains experimental for long-horizon autonomous debugging.
• Vision inputs are unsupported.

License

This model inherits the upstream Qwen/Qwen3.5-9B Apache-2.0 license. See LICENSE.