Alibaba (Qwen) · Released Aug 14, 2026Qwen3.8 27B
Dense 27B vision-language model for coding, agent tasks, and image and video understanding.
- Capability
- 149.4ECI · #53 of 148
- Input
- $0.40per 1M tokens
- Output
- $2.50per 1M tokens
- Context
- 262K33K max output
What it costs
- Input
- $0.40per million tokens
- Output
- $2.50per million tokens
- Blended (3:1)
- $0.925pricier than 53% of priced models
Typical monthly bills
| Monthly usage | Estimated cost |
|---|---|
| Side project2M input + 0.5M output tokens | $2.05 |
| Team assistant25M input + 5M output tokens | $22.50 |
| Production app250M input + 50M output tokens | $225.00 |
Median of 39 paid API providers on models.dev.
Independent benchmarks
#53 of 148 scored models
The shaded band is Epoch AI’s confidence range (147.5–151.6); the tick marks the median scored model.
Scores from Epoch AI, run independently of Alibaba (Qwen).
The details
- Lab
- Alibaba (Qwen)
- Released
- Aug 14, 2026
- Context window
- 262,144 tokens
- Max output
- 32,768 tokens
- Inputs
- Text, Images, Video
- Output
- Text
- Reasoning
- Always on
- Tool calling
- Yes
- Structured output
- Yes
- Weights
- Open weights
- Availability
- 41 API providerslisted on models.dev
What Alibaba (Qwen) claims
Published by the lab at launch. Settings vary, so compare these only with care.
| Benchmark | Score | Setting | Source |
|---|---|---|---|
| SWE-bench Pro | 61.7 resolved | — | Source |
| Terminal-Bench v2.1 | 73 | — | Source |
| NL2Repo | 42.3 | — | Source |
| DeepSWE v1.1 | 42.2 | — | Source |
| QwenSWEBench | 79 avg@3 | — | Source |
| CoWorkBench | 70.7 | — | Source |
| JobBench | 33.4 | — | Source |
| Agents' Last Exam | 20.4 | — | Source |
| Agents' Last Exam | 42.9 | — | Source |
| IFBench | 79.5 | — | Source |
| GPQA Diamond | 89.2 | — | Source |
| Humanity's Last Exam | 30.8 | without tools; GPT-4o judge | Source |
More from Alibaba (Qwen)
About Qwen3.8 27B
How much does Qwen3.8 27B cost?
Qwen3.8 27B costs $0.40 input / $2.50 output per million tokens (median across 39 API providers). At a 3:1 input-to-output mix that is $0.925 per million tokens, more expensive than 53% of the 360 priced models we track.
What is the context window of Qwen3.8 27B?
Qwen3.8 27B accepts up to 262,144 tokens per request and can write up to 32,768 tokens in one response.
How good is Qwen3.8 27B?
Epoch AI gives Qwen3.8 27B a Capabilities Index score of 149.4 (likely range 147.5–151.6), ranking it #53 of 148 models Epoch has scored.
Is Qwen3.8 27B open source?
Yes. Alibaba (Qwen) publishes the weights, so it can be downloaded and self-hosted.
What inputs does Qwen3.8 27B support?
Qwen3.8 27B accepts text, images and video and replies in text. It is a reasoning model, supports tool calling and can return structured JSON output.
When was Qwen3.8 27B released?
Alibaba (Qwen) released Qwen3.8 27B on Aug 14, 2026.
What are the best alternatives to Qwen3.8 27B?
The closest current models from other labs on capability, price and release date are Inkling Small (Thinking Machines, ECI 150.2, $0.50 / $1.20), Kimi K2.7 Code (Moonshot AI, ECI 150.0, $0.95 / $4.00), Nemotron 3 Ultra 550B A55B (NVIDIA, ECI 146.2, $0.50 / $2.50) and Grok 4.3 (xAI, ECI 149.2, $1.25 / $2.50).