Qwen3.5-397B-A17B
Released2026-02-16Size397B · 17B activeArchitecturehybrid linear attnLicenseapache-2.0AA rank#31 of 75
Files and memory at 8K context
| Quant | File | Files | + KV + runtime | Smallest fit | Source |
|---|---|---|---|---|---|
| Q4_K_M unsloth/Qwen3.5-397B-A17B-GGUF | 227 GiB | 6 | 229 GiB+ | over 192 GB GPU | HF files |
| Q5_K_M unsloth/Qwen3.5-397B-A17B-GGUF | 273 GiB | 8 | 275 GiB+ | over 192 GB GPU | HF files |
| Q8_0 unsloth/Qwen3.5-397B-A17B-GGUF | 393 GiB | 10 | 394 GiB+ | over 192 GB GPU | HF files |
| BF16 unsloth/Qwen3.5-397B-A17B-GGUF | 739 GiB | 17 | 740 GiB+ | over 192 GB GPU | HF files |
| MLX 4-bit mlx-community/Qwen3.5-397B-A17B-4bit | 208 GiB | 46 | 210 GiB+ | 512 GB Mac | HF files |
| MLX 8-bit mlx-community/Qwen3.5-397B-A17B-8bit | 393 GiB | 91 | 394 GiB+ | over 512 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
21.4 AA estimate
Variant: Qwen3.5 397B A17B (Non-reasoning) · observed 2026-10-17 · source
Rank #31 of 75 researched 2026 models.
Arena text rating
1,438
95% CI 1435–1441 · 93,567 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 0
No reported measurement yet. Submit one.