Alibaba
Qwen3.5 122B-A10B
MoE
125,1B
Parameters · 10B active
72,3 Go
Q4_K_M file
≥ 73,8 Go
Recommended VRAM
262 144
Max context
Not measured yet
No card has run the TokenBench protocol on this model yet. The current protocol covers a selection of models; benchmarking any catalog model from the app is coming soon.
Measure my card →Which cards should it fit on?
Cards on the site whose video memory covers the recommended 73,8 Go (file + 4,000-token context). An estimate, not a measurement. Click a card for the detailed estimate.
Technical sheet
- Published
- 2026-02-24
- Architecture
- 48 layers · 2 KV heads · dimension 256
- Attention
- Hybrid: 12 of 48 layers use standard attention, the others a fixed-memory mechanism (Mamba, linear attention) — a very light cache.
- Context cache (f16)
- ≈ 23 MB per 1,000 tokens · 0,8 Go for 32,768 tokens
- GGUF file
- bartowski/Qwen_Qwen3.5-122B-A10B-GGUF · 2 files Download (72,3 Go)
- Official repo
- Qwen/Qwen3.5-122B-A10B
Public Hugging Face data (official config.json, GGUF repo). The context cache only counts full-attention layers.