Qwen instruction model for multilingual chat, reasoning, and tool use
Specification
most-agreed values
Context
33K
Max output
16K
Released
2025-01-01
Knowledge cutoff
—
Retires
—
Open weights
no
Input
text
Output
text
Price
US dollars per million tokens · most-agreed
Input
free
Output
free
Quality
3 benchmarks · 1 source
Composite
17th
percentile of 334 scored models
Rank
277 / 334
±0.10 sd
Evidence
3 × 1
one source, or fewer than 3 benchmarks
Effort range
—
one configuration only
reasoning15th2/4 bench
math19th1/7 bench
Every score
one row per source and configuration — nothing averaged away
| Benchmark | Score | Configuration | Source | Run |
|---|---|---|---|---|
| OTIS Mock AIME 2024-2025 | 21.4 ±5.3 | default | Epoch AI | 2026-08-28 |
| Chess Puzzles | 0.0 ±0.0 | default | Epoch AI | 2026-08-28 |
| GPQA diamond | 33.6 ±2.1 | default | Epoch AI | 2026-08-28 |
The leading figure for a benchmark is the median across its configurations, so one heroic high-effort run cannot set the number.
Available from 4 hosts
| Host | In $/M | Out $/M | Cache rd | Cache wr | Context | Output | Retires |
|---|---|---|---|---|---|---|---|
| alibaba-cn | free | free | — | — | 33K | 16K | — |
| nscale | $0.09 | $0.09 | — | — | — | — | — |
| fireworks_ai | $0.10 | $0.10 | — | — | 131K | 131K | — |
| together_ai | $0.18 | $0.18 | — | — | — | — | — |
More from DeepSeek
most-hosted first