- input
- $0.26
- output
- $2.08
- Context
- 262Ktokens
- Released
- 8mo ago
- Size
- 122Bparameters
Cost/1M tokens
Open-weight builds
Ollama variants
Every published Ollama tag matched to this model, including parameter editions, instruction/base variants, and quantized builds.
| Variant | Parameters | Type | Quantization | Size | Context | Input |
|---|---|---|---|---|---|---|
| qwen3.5:0.8b-bf16 | 0.8B | — | BF16 | 1.8 GB | 256K | text + image |
| qwen3.5:0.8b-mlx-bf16 | 0.8B | — | BF16 | 1.7 GB | 256K | text + image |
| qwen3.5:0.8b-mtp-q8_0 | 0.8B | — | Q8_0 | 1.3 GB | 256K | text + image |
| qwen3.5:0.8b-q8_0 | 0.8B | — | Q8_0 | 1 GB | 256K | text + image |
| qwen3.5:0.8b | 0.8B | — | — | 1.2 GB | 256K | text + image |
| qwen3.5:0.8b-mlx | 0.8B | — | — | 1.2 GB | 256K | text + image |
| qwen3.5:0.8b-nvfp4 | 0.8B | — | — | 1 GB | 256K | text + image |
| qwen3.5:2b-bf16 | 2B | — | BF16 | 4.6 GB | 256K | text + image |
| qwen3.5:2b-mlx-bf16 | 2B | — | BF16 | 4.4 GB | 256K | text + image |
| qwen3.5:2b-mtp-q4_K_M | 2B | — | Q4_K_M | 2.6 GB | 256K | text + image |
| qwen3.5:2b-q4_K_M | 2B | — | Q4_K_M | 1.9 GB | 256K | text + image |
| qwen3.5:2b-mtp-q8_0 | 2B | — | Q8_0 | 3.3 GB | 256K | text + image |
| qwen3.5:2b-q8_0 | 2B | — | Q8_0 | 2.7 GB | 256K | text + image |
| qwen3.5:2b | 2B | — | — | 2.7 GB | 256K | text + image |
| qwen3.5:2b-mlx | 2B | — | — | 3.1 GB | 256K | text + image |
| qwen3.5:2b-nvfp4 | 2B | — | — | 2.5 GB | 256K | text + image |
| qwen3.5:4b-bf16 | 4B | — | BF16 | 9.3 GB | 256K | text + image |
| qwen3.5:4b-mlx-bf16 | 4B | — | BF16 | 9.1 GB | 256K | text + image |
| qwen3.5:4b-mtp-q4_K_M | 4B | — | Q4_K_M | 4.1 GB | 256K | text + image |
| qwen3.5:4b-q4_K_M | 4B | — | Q4_K_M | 3.3 GB | 256K | text + image |
| qwen3.5:4b-mtp-q8_0 | 4B | — | Q8_0 | 6 GB | 256K | text + image |
| qwen3.5:4b-q8_0 | 4B | — | Q8_0 | 5.2 GB | 256K | text + image |
| qwen3.5:4b | 4B | — | — | 3.3 GB | 256K | text + image |
| qwen3.5:4b-mlx | 4B | — | — | 4 GB | 256K | text + image |
| qwen3.5:4b-mxfp8 | 4B | — | — | 5.6 GB | 256K | text + image |
| qwen3.5:9b-bf16 | 9B | — | BF16 | 19 GB | 256K | text + image |
| qwen3.5:9b-mlx-bf16 | 9B | — | BF16 | 19 GB | 256K | text + image |
| qwen3.5:9b-mtp-q4_K_M | 9B | — | Q4_K_M | 9 GB | 256K | text + image |
| qwen3.5:9b-q4_K_M | 9B | — | Q4_K_M | 6.6 GB | 256K | text + image |
| qwen3.5:9b-mtp-q8_0 | 9B | — | Q8_0 | 13 GB | 256K | text + image |
| qwen3.5:9b-q8_0 | 9B | — | Q8_0 | 10 GB | 256K | text + image |
| qwen3.5:9b-mlx | 9B | — | — | 8.9 GB | 256K | text + image |
| qwen3.5:9b-mtp-nvfp4 | 9B | — | — | 7.6 GB | 256K | text + image |
| qwen3.5:9b-mxfp8 | 9B | — | — | 11 GB | 256K | text + image |
| qwen3.5:9b-nvfp4 | 9B | — | — | 7.2 GB | 256K | text + image |
| qwen3.5:latest | 9B | — | — | 6.6 GB | 256K | text + image |
| qwen3.5:27b-bf16 | 27B | — | BF16 | 56 GB | 256K | text + image |
| qwen3.5:27b-coding-bf16 | 27B | — | BF16 | 55 GB | 256K | text + image |
| qwen3.5:27b-mlx-bf16 | 27B | — | BF16 | 55 GB | 256K | text + image |
| qwen3.5:27b-mtp-q4_K_M | 27B | — | Q4_K_M | 21 GB | 256K | text + image |
| qwen3.5:27b-q4_K_M | 27B | — | Q4_K_M | 17 GB | 256K | text + image |
| qwen3.5:27b-q8_0 | 27B | — | Q8_0 | 30 GB | 256K | text + image |
| qwen3.5:27b | 27B | — | — | 17 GB | 256K | text + image |
| qwen3.5:27b-coding-mxfp8 | 27B | — | — | 30 GB | 256K | text + image |
| qwen3.5:27b-coding-nvfp4 | 27B | — | — | 20 GB | 256K | text + image |
| qwen3.5:27b-int4 | 27B | — | — | 16 GB | 256K | text + image |
| qwen3.5:27b-int8 | 27B | — | — | 30 GB | 256K | text + image |
| qwen3.5:27b-mlx | 27B | — | — | 20 GB | 256K | text + image |
| qwen3.5:27b-mxfp8 | 27B | — | — | 30 GB | 256K | text + image |
| qwen3.5:35b-a3b-bf16 | 35B | — | BF16 | 72 GB | 256K | text + image |
| qwen3.5:35b-a3b-coding-bf16 | 35B | — | BF16 | 70 GB | 256K | text + image |
| qwen3.5:35b-a3b-mlx-bf16 | 35B | — | BF16 | 70 GB | 256K | text + image |
| qwen3.5:35b-a3b-mtp-q4_K_M | 35B | — | Q4_K_M | 24 GB | 256K | text + image |
| qwen3.5:35b-a3b-q4_K_M | 35B | — | Q4_K_M | 22 GB | 256K | text + image |
| qwen3.5:35b-a3b-q8_0 | 35B | — | Q8_0 | 39 GB | 256K | text + image |
| qwen3.5:35b | 35B | — | — | 22 GB | 256K | text + image |
| qwen3.5:35b-a3b-coding-mxfp8 | 35B | — | — | 37 GB | 256K | text + image |
| qwen3.5:35b-a3b-coding-nvfp4 | 35B | — | — | 22 GB | 256K | text + image |
| qwen3.5:35b-a3b-int4 | 35B | — | — | 20 GB | 256K | text + image |
| qwen3.5:35b-a3b-int8 | 35B | — | — | 38 GB | 256K | text + image |
| qwen3.5:35b-a3b-mxfp8 | 35B | — | — | 37 GB | 256K | text + image |
| qwen3.5:35b-a3b-nvfp4 | 35B | — | — | 22 GB | 256K | text + image |
| qwen3.5:122b | 122B | — | Q4_K_M | 81 GB | 256K | text + image |
Deployment graph
Can I run this?
Vertical Tabs only shows compatibility when it has structured evidence tying this exact model to hardware, runtime, quantization, memory use, and observed performance.