- input
- $0.21
- cached
- $0.10
- output
- $1.90
- Context
- 262Ktokens
- Released
- 1y ago
- Size
- 235Bparameters
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-vl:2b-instruct-bf16 | 2B | instruct | BF16 | 4.3 GB | 256K | text + image |
| qwen3-vl:2b-instruct | 2B | instruct | Q4_K_M | 1.9 GB | 256K | text + image |
| qwen3-vl:2b-instruct-q8_0 | 2B | instruct | Q8_0 | 2.6 GB | 256K | text + image |
| qwen3-vl:2b-thinking-bf16 | 2B | thinking | BF16 | 4.3 GB | 256K | text + image |
| qwen3-vl:2b | 2B | thinking | Q4_K_M | 1.9 GB | 256K | text + image |
| qwen3-vl:2b-thinking-q8_0 | 2B | thinking | Q8_0 | 2.6 GB | 256K | text + image |
| qwen3-vl:4b-instruct-bf16 | 4B | instruct | BF16 | 8.9 GB | 256K | text + image |
| qwen3-vl:4b-instruct | 4B | instruct | Q4_K_M | 3.3 GB | 256K | text + image |
| qwen3-vl:4b-instruct-q8_0 | 4B | instruct | Q8_0 | 5.1 GB | 256K | text + image |
| qwen3-vl:4b-thinking-bf16 | 4B | thinking | BF16 | 8.9 GB | 256K | text + image |
| qwen3-vl:4b | 4B | thinking | Q4_K_M | 3.3 GB | 256K | text + image |
| qwen3-vl:4b-thinking-q8_0 | 4B | thinking | Q8_0 | 5.1 GB | 256K | text + image |
| qwen3-vl:8b-instruct-bf16 | 8B | instruct | BF16 | 18 GB | 256K | text + image |
| qwen3-vl:8b-instruct | 8B | instruct | Q4_K_M | 6.1 GB | 256K | text + image |
| qwen3-vl:8b-instruct-q8_0 | 8B | instruct | Q8_0 | 9.8 GB | 256K | text + image |
| qwen3-vl:8b-thinking-bf16 | 8B | thinking | BF16 | 18 GB | 256K | text + image |
| qwen3-vl:latest | 8B | thinking | Q4_K_M | 6.1 GB | 256K | text + image |
| qwen3-vl:8b-thinking-q8_0 | 8B | thinking | Q8_0 | 9.8 GB | 256K | text + image |
| qwen3-vl:30b-a3b-instruct-bf16 | 30B | instruct | BF16 | 62 GB | 256K | text + image |
| qwen3-vl:30b-a3b-instruct | 30B | instruct | Q4_K_M | 20 GB | 256K | text + image |
| qwen3-vl:30b-a3b-instruct-q8_0 | 30B | instruct | Q8_0 | 34 GB | 256K | text + image |
| qwen3-vl:30b-a3b-thinking-bf16 | 30B | thinking | BF16 | 62 GB | 256K | text + image |
| qwen3-vl:30b | 30B | thinking | Q4_K_M | 20 GB | 256K | text + image |
| qwen3-vl:30b-a3b-thinking-q8_0 | 30B | thinking | Q8_0 | 34 GB | 256K | text + image |
| qwen3-vl:32b-instruct-bf16 | 32B | instruct | BF16 | 67 GB | 256K | text + image |
| qwen3-vl:32b-instruct | 32B | instruct | Q4_K_M | 21 GB | 256K | text + image |
| qwen3-vl:32b-instruct-q8_0 | 32B | instruct | Q8_0 | 36 GB | 256K | text + image |
| qwen3-vl:32b-thinking-bf16 | 32B | thinking | BF16 | 67 GB | 256K | text + image |
| qwen3-vl:32b | 32B | thinking | Q4_K_M | 21 GB | 256K | text + image |
| qwen3-vl:32b-thinking-q8_0 | 32B | thinking | Q8_0 | 36 GB | 256K | text + image |
| qwen3-vl:235b-a22b-instruct-bf16 | 235B | instruct | BF16 | 471 GB | 256K | text + image |
| qwen3-vl:235b-a22b-instruct | 235B | instruct | Q4_K_M | 143 GB | 256K | text + image |
| qwen3-vl:235b-a22b-instruct-q8_0 | 235B | instruct | Q8_0 | 251 GB | 256K | text + image |
| qwen3-vl:235b-a22b-thinking-bf16 | 235B | thinking | BF16 | 471 GB | 256K | text + image |
| qwen3-vl:235b | 235B | thinking | Q4_K_M | 143 GB | 256K | text + image |
| qwen3-vl:235b-a22b-thinking-q8_0 | 235B | thinking | Q8_0 | 251 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.