All models
Alibaba (Qwen)
Qwen3 VL 8B Instruct fastruns locally
alibaba/qwen3-vl-8b
Qwen3-VL-8B-Instruct is a multimodal vision-language model from the Qwen3-VL series, built for high-fidelity understanding and reasoning across text, images, and video. It features improved multimodal fusion with Interleaved-MRoPE for long-horizon...
Input / 1M
$0.25
USD per million tokens
Output / 1M
$0.75
USD per million tokens
Est. cost / token
$0.000000375
$0.38 per 1M, 3:1 mix
Context window
262K
max output 236K
Major features
VisionOpen weightsCachingJSON schemaTools
Task suitability (heuristic 1–5, not a benchmark)
- coding3
- reasoning2
- math2
- summarization4
- extraction4
- classification4
- chat3
- creative3
- translation4
- vision3
- long context3
- agentic3
Run it locally
Will it run on my machine? →ollama run qwen3:8b| Quantization | Quality | Memory needed (8K ctx) |
|---|---|---|
| Q2_K | poor | 5.6 GB |
| Q3_K_M | fair | 6.2 GB |
| Q4_K_Mrecommended | good | 7.4 GB |
| Q5_K_M | very good | 8.3 GB |
| Q6_K | excellent | 9.3 GB |
| Q8_0 | near-lossless | 11.5 GB |
| F16 | lossless | 20.2 GB |
Estimates.
Price history
| Recorded | Input | Output | Cached | Source |
|---|---|---|---|---|
| 2026-11-05 03:20 | $0.25 | $0.75 | $0.12 | openrouter |
A row is added whenever a sync detects a price change. Vendor pricing source
Pricing (USD per 1M tokens)
- Input
- $0.25
- Output
- $0.75
- Cached input
- $0.12
- Blended (3:1)
- $0.38
Details
- Company
- Alibaba (Qwen)
- Version
- 3
- Family
- Qwen
- Released
- 2025-10-14
- Parameters
- 8.8B
- Max output
- 236K
- Open weights
- Yes
- Inputs
- image, text
- Last synced
- 25 min ago