Qwen3.5-0.8B
Files and memory at 8K context
| Quant | File | Files | + KV + runtime | Smallest fit | Source |
|---|---|---|---|---|---|
| Q4_K_M unsloth/Qwen3.5-0.8B-GGUF | 0.50 GiB | 1 | 1.6 GiB+ | 8 GB GPU | HF files |
| Q5_K_M unsloth/Qwen3.5-0.8B-GGUF | 0.55 GiB | 1 | 1.6 GiB+ | 8 GB GPU | HF files |
| Q8_0 unsloth/Qwen3.5-0.8B-GGUF | 0.76 GiB | 1 | 1.8 GiB+ | 8 GB GPU | HF files |
| FP16 mradermacher/Qwen3.5-0.8B-GGUF | 1.41 GiB | 1 | 2.5 GiB+ | 8 GB GPU | HF files |
| MLX 4-bit mlx-community/Qwen3.5-0.8B-4bit | 0.58 GiB | 1 | 1.7 GiB+ | 8 GB Mac | HF files |
| MLX 8-bit mlx-community/Qwen3.5-0.8B-8bit | 0.93 GiB | 1 | 2.0 GiB+ | 8 GB Mac | HF files |
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
Unknown
Arena text rating
Unknown
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.
Community threads
- 0.8B with mtp is slower than without mtp Hugging Face · 2026-04-09
- When vllm deployment is started using mtp config, an error will be reported. Please help！ Hugging Face · 2026-04-03
Qualitative signals only; titles are not verified claims.