Qwen/Qwen-AgentWorld-35B-A3B
Full fine-tune (fewer layers) vs Qwen/Qwen3.5-35B-A3B-Base. Qwen/Qwen-AgentWorld-35B-A3B matches part of Qwen/Qwen3.5-35B-A3B-Base (693 of its 1811 tensors; the rest of Qwen/Qwen3.5-35B-A3B-Base has no counterpart). On the shared part: Fine-tune: 100% of the weights changed, across 40 of 40 layers.
Tensors changed: MLP 280/280, lm_head 1/1, Embeddings 1/1, Attention 272/280, Norms 131/131
26 derivatives checked
- unsloth/Qwen-AgentWorld-35B-A3B-GGUF: Same weights, converted
- SC117/Qwen-AgentWorld-35B-A3B-MTP-Uncensored-APEX-GGUF: Abliterated
- DhruvalLabs/Qwen-AgentWorld-35B-A3B-GGUF: Quantization
- SC117/Qwen-AgentWorld-35B-A3B-MTP-APEX-GGUF: Quantization
- cyankiwi/Qwen-AgentWorld-35B-A3B-AWQ-INT4: Quantization
- DevQuasar/Qwen.Qwen-AgentWorld-35B-A3B-GGUF: Same weights, converted (fewer layers)
- mudler/Qwen-AgentWorld-35B-A3B-APEX-GGUF: Quantization
- mlx-community/Qwen-AgentWorld-35B-A3B-oQ4: Quantization
- Jiunsong/SuperQwen-AgentWorld-35B-A3B-abliterated: LoRA-style
- vadikulous-ai/Qwen-AgentWorld-35B-A3B-UD-Q4_K_XL-MTP-GGUF: Quantization
- groxaxo/Qwen-AgentWorld-35B-A3B-GGUF: Same weights, converted
- AtomicChat/qwen-agentworld-35b-GGUF: Quantization
- cyankiwi/Qwen-AgentWorld-35B-A3B-AWQ-FP8: Quantization
- lovedheart/Qwen-AgentWorld-35B-A3B-NVFP4: Quantization
- huihui-ai/Huihui-Qwen-AgentWorld-35B-A3B-abliterated: Abliterated
- hotdogs/Qwen35B-Agent-R2: LoRA-style
- FreedomAISVR/Qwen-AgentWorld-35B-A3B-MXFP4-MOE-GGUF: Quantization
- lovedheart/Qwen-AgentWorld-35B-A3B-FP8: Full fine-tune
- sakamakismile/Qwen-AgentWorld-35B-A3B-NVFP4: Quantization
- groxaxo/Qwen-AgentWorld-35B-A3B-GPTQ-Pro-Int4: Quantization
- Brooooooklyn/Qwen-AgentWorld-35B-A3B-mxfp4-mlx: Quantization
- DreamFoundries/Qwen-AgentWorld-35B-A3B-6bit: Quantization
- FreedomAISVR/Qwen-AgentWorld-35B-A3B-NVFP4-GGUF: Quantization
- jedisct1/Qwen-AgentWorld-35B-A3B-REAP78-oQ4-MLX: Quantization
- Frosty40/Qwen-AgentWorld-35B-A3B-B70-Turbo-GGUF: Quantization
- YOYO-AI/Qwen3.6-35B-A3B-YOYO-V2: Full fine-tune (fewer layers)
9 other repos hold byte-for-byte identical weight files.