Qwen3.5-122B-A10B
Released2026-02-24Size122B · 10B activeArchitecturehybrid linear attnLicenseapache-2.0AA rank#37 of 75
Files and memory at 8K context
| Quant | File | Files | + KV + runtime | Smallest fit | Source |
|---|---|---|---|---|---|
| Q4_K_M unsloth/Qwen3.5-122B-A10B-GGUF | 71.28 GiB | 3 | 72.5 GiB+ | 80 GB GPU | HF files |
| Q5_K_M unsloth/Qwen3.5-122B-A10B-GGUF | 85.23 GiB | 3 | 86.4 GiB+ | 96 GB GPU | HF files |
| Q8_0 unsloth/Qwen3.5-122B-A10B-GGUF | 121 GiB | 4 | 122 GiB+ | 128 GB GPU | HF files |
| BF16 unsloth/Qwen3.5-122B-A10B-GGUF | 228 GiB | 5 | 229 GiB+ | over 192 GB GPU | HF files |
| MLX 4-bit mlx-community/Qwen3.5-122B-A10B-4bit | 64.81 GiB | 14 | 66.0 GiB+ | 96 GB Mac | HF files |
| MLX 8-bit mlx-community/Qwen3.5-122B-A10B-8bit | 122 GiB | 25 | 123 GiB+ | 192 GB Mac | HF files |
No verified file: FP16. Sizes are exact file bytes (text weights; projectors listed separately). “+” = recurrent state or other terms unknown. Smallest fit adds 0.5 GiB spare; Macs use 75% of memory.
Quality
Artificial Analysis Intelligence Index
17.7 AA estimate
Variant: Qwen3.5 122B A10B (Non-reasoning) · observed 2026-10-17 · source
Rank #37 of 75 researched 2026 models.
Arena text rating
1,417
95% CI 1413–1422 · 29,732 votes · published 2026-10-15 · dataset
Exact listed model name; hosted reasoning/quantization may differ from local checkpoint.
Scores describe the evaluated hosted variant, not each quantized file. AA and Arena are never averaged.
Reported speeds 10
Reported by the linked sources, not reproduced by LocalFit. All rows in the speed database