havok2

Qwen-AgentWorld-35B-A3B-VL36

README

License: apache-2.0

What we changed vs. Qwen-AgentWorld-35B-A3B

AgentWorld ships text-only weights (693 tensors, all model.language_model.* + lm_head) even though its config already declares a vision_config, image/video token ids, and a vision+tool-capable chat template — it is effectively a Qwen3.5-VL waiting for a vision encoder. We:

Extracted the 333 model.visual.* tensors (27-layer ViT + patch/position embeds + merger/projector) from Qwen3.6-35B-A3B (BF16).
Added them to the AgentWorld checkpoint without touching a single text tensor — so the agent / tool-calling skills are byte-for-byte preserved.
Removed the language_model_only flag from the config so the runner instantiates the tower.

Result: 1026 tensors (693 text + 333 vision), architecture Qwen3_5MoeForConditionalGeneration, projector maps vision hidden 1152 → text hidden 2048.

Serving (vLLM ≥ 0.23)

bash
vllm serve havok2/Qwen-AgentWorld-35B-A3B-VL36 \
  --tensor-parallel-size 4 \
  --reasoning-parser qwen3 \
  --limit-mm-per-prompt '{"image": 4, "video": 1}'
# IMPORTANT: do NOT pass --language-model-only — that disables the vision tower.

Loads as a standard Qwen3.5-MoE VLM. ~70 GB BF16; fits on 4×24 GB with --tensor-parallel-size 4.

Capabilities & limitations

✅ Agentic text + tool-calling — identical to AgentWorld (weights untouched).
✅ Image / video input — works; smoke-tested on synthetic images (shapes, colors, OCR).
⚠️ Vision quality is not benchmarked; cross-generation graft, may differ from a natively co-trained VLM.
⚠️ No additional safety tuning or evaluation beyond basic smoke tests.

Reproduce it yourself (no 70 GB download needed)

Rebuild locally from the two source models with the included build_merged_vl36.py. It downloads the AgentWorld backbone and only the Qwen3.6 shards that contain the vision tensors, then performs the graft:

bash
pip install huggingface_hub safetensors torch
python build_merged_vl36.py        # -> ./Qwen-AgentWorld-35B-A3B-VL36

Attribution & license

Text backbone & agent post-training: Qwen/Qwen-AgentWorld-35B-A3B (Apache-2.0), itself based on Qwen3.5-35B-A3B-Base.
Vision tower (ViT + projector): Qwen/Qwen3.6-35B-A3B (Apache-2.0).

This derivative is released under Apache-2.0. Modification: vision-tower graft as described above (no training). See LICENSE. Not affiliated with or endorsed by Alibaba / the Qwen team.

Available on FriendliAI

Dedicated Endpoints

Run this model inference on single tenant GPU with unmatched speed and reliability at scale.

Learn more

Model Details

Model Provider

havok2

Model Tree

Base

Qwen/Qwen-AgentWorld-35B-A3B

Base

Qwen/Qwen3.6-35B-A3B

Merged

this model

Input Modalities

TextImageVideo

Output Modalities

Text

Supported Functionality