Qwen vision-language model for visual reasoning, documents, and agent tasks
Specification
alibaba listing
Context
1M
Max output
66K
Released
2026-02-16
Knowledge cutoff
2025-04
Retires
—
Open weights
no
Input
text, image, video
Output
text
Price
US dollars per million tokens · alibaba list price
Input
$0.40
Output
$2.40
Cache read
$0.04
Cache write
$0.375
Long-context price tiers
the rate above each context threshold
| Above | Price | $/M |
|---|---|---|
| 256K | Cache write | $0.4688 |
| 256K | Input | $0.375 |
| 256K | Output | $2.25 |
Quality
7 benchmarks · 1 source
Composite
65th
percentile of 334 scored models
Rank
119 / 334
±0.07 sd
Evidence
7 × 1
one source, or fewer than 3 benchmarks
Effort range
—
one configuration only
reasoning74th3/4 bench
math74th3/7 bench
knowledge21st1/1 bench
Every score
one row per source and configuration — nothing averaged away
| Benchmark | Score | Configuration | Source | Run |
|---|---|---|---|---|
| OTIS Mock AIME 2024-2025 | 86.7 ±5.1 | default | Epoch AI | 2026-08-07 |
| GPQA diamond | 84.8 ±2.6 | default | Epoch AI | 2026-08-07 |
| FrontierMath-v1 | 35.5 ±2.4 | default · 2 runs | Epoch AI | 2026-05-14 |
| Chess Puzzles | 22.0 ±4.2 | default | Epoch AI | 2026-08-07 |
| Mystery Game Puzzles | 16.0 ±3.7 | default | Epoch AI | 2026-08-05 |
| FrontierMath-Tier-4-2025-07-01 | 2.1 ±2.1 | default | Epoch AI | 2026-05-15 |
| SimpleQA Verified | 25.4 ±1.4 | default | Epoch AI | 2026-08-27 |
The leading figure for a benchmark is the median across its configurations, so one heroic high-effort run cannot set the number.
Available from 23 hosts
| Host | In $/M | Out $/M | Cache rd | Cache wr | Context | Output | Retires |
|---|---|---|---|---|---|---|---|
| alibabavendor | $0.40 | $2.40 | — | — | 1M | 66K | — |
| alibaba-coding-plan | free | free | free | free | 1M | 66K | — |
| alibaba-coding-plan-cn | free | free | free | free | 1M | 66K | — |
| merge-gateway | $0.115 | $0.688 | $0.023 | — | 1M | 250K | — |
| orcarouter | $0.115 | $0.688 | — | — | 1M | 66K | — |
| 302ai | $0.12 | $0.69 | — | — | 1M | 66K | — |
| opencode | $0.20 | $1.20 | $0.02 | $0.25 | 262K | 66K | — |
| opencode-go | $0.20 | $1.20 | $0.02 | $0.25 | 262K | 66K | — |
| kilo | $0.30 | $1.80 | — | $0.375 | 1M | 66K | — |
| openrouter | $0.30 | $1.80 | — | $0.375 | 1M | 66K | — |
| empiriolabs | $0.36 | $2.21 | $0.36 | — | 1M | 66K | — |
| llmtr | $0.40 | $2.40 | — | — | 1M | 66K | — |
| meganova | $0.40 | $2.40 | — | — | 1M | 66K | — |
| nano-gpt | $0.40 | $2.40 | $0.04 | — | 984K | 66K | — |
| nano-gpt · thinking | $0.40 | $2.40 | $0.04 | — | 984K | 66K | — |
| ofox | $0.40 | $2.40 | $0.04 | $0.40 | 1M | 64K | — |
| ofox · bailian | $0.40 | $2.40 | $0.04 | $0.40 | 1M | 66K | — |
| vercel | $0.40 | $2.50 | $0.04 | $0.50 | 1M | 64K | — |
| alibaba-cn | $0.573 | $3.44 | — | — | 1M | 66K | — |
| zenmux | $0.80 | $4.80 | — | — | 1M | 64K | — |
| dashscope · dashscope | — | — | — | — | 992K | 66K | — |
| qwen_ai_platform · qwen-ai-platform | — | — | — | — | 992K | 66K | — |
| qwencloud · qwencloud | — | — | — | — | 992K | 66K | — |
Price history
append-only observations · a listing writes a row only when its price moves
| Date | Host | Field | $/M |
|---|---|---|---|
| 2026-09-02 | openrouter | Cache write | $0.375 |
| 2026-09-06 | openrouter | Cache write | $0.375 |
| 2026-09-02 | openrouter | Input | $0.30 |
| 2026-09-06 | openrouter | Input | $0.30 |
| 2026-09-02 | openrouter | Output | $1.80 |
| 2026-09-06 | openrouter | Output | $1.80 |
| 2026-09-02 | vercel | Output | $2.40 |
| 2026-09-15 | vercel | Output | $2.50 |
More from Alibaba
most-hosted first