# Qwen3.8 27B Usage, Cost & Rank | OpenCode Data

> Qwen3.8 27B ranked #46 by tokens across OpenCode last week, with 0% of tokens over the past two months. Qwen3.8 27B costs $0.50 per 1M input tokens and $3.00 per 1M output tokens.

Dense 27B vision-language model for coding, agent tasks, and image and video understanding

- Page: https://opencode.ai/data/alibaba/qwen3-8-27b
- JSON: https://opencode.ai/data/alibaba/qwen3-8-27b.json
- Model ID: alibaba/qwen3.8-27b
- Lab: Alibaba
- Updated: 2026-10-02T08:39:37.000Z

## Model facts

| Fact | Value |
| --- | --- |
| Context window | 262K |
| Max output | 33K |
| Knowledge cutoff | - |
| Release date | 2026-08-14 |
| Input modalities | text, image, video |
| Output modalities | text |
| Reasoning | Yes |
| Tool calling | Yes |
| Open weights | Yes |
| Weights | [Hugging Face](https://huggingface.co/Qwen/Qwen3.8-27B) |

## Pricing: USD per 1M tokens

| Input | Output | Cached input | Cache write |
| --- | --- | --- | --- |
| $0.50 | $3.00 | $0.10 | $0.63 |

## OpenCode usage: past 2 months

| Metric | Value |
| --- | --- |
| Rank by tokens last week | #46 |
| Tokens | 186M |
| Share of all tokens | 0% |
| Change vs previous 2 months | +100% |
| Unique users | 52 |
| Completed sessions | 318 |
| Average tokens per session | 585K |
| Average cost per session | $0.0000 |
| Total spend | $0 |
| Input tokens served from cache | 88.6% |
| Weekly retention | - |

## Daily usage: past 2 months

| Date | Tokens | Unique users | Sessions |
| --- | --- | --- | --- |
| 2026-08-26 | 186M | 50 | 314 |
| 2026-08-27 | 0 | 0 | 0 |
| 2026-08-28 | 0 | 0 | 0 |
| 2026-08-29 | 0 | 0 | 0 |
| 2026-08-30 | 0 | 0 | 0 |
| 2026-08-31 | 0 | 0 | 0 |
| 2026-09-01 | 0 | 0 | 0 |
| 2026-09-02 | 0 | 0 | 0 |
| 2026-09-03 | 0 | 0 | 0 |
| 2026-09-04 | 0 | 0 | 0 |
| 2026-09-05 | 0 | 0 | 0 |
| 2026-09-06 | 0 | 0 | 0 |
| 2026-09-07 | 0 | 0 | 0 |
| 2026-09-08 | 0 | 0 | 0 |
| 2026-09-09 | 0 | 0 | 0 |
| 2026-09-10 | 0 | 0 | 0 |
| 2026-09-11 | 0 | 0 | 0 |
| 2026-09-12 | 0 | 0 | 0 |
| 2026-09-13 | 0 | 0 | 0 |
| 2026-09-14 | 0 | 0 | 0 |
| 2026-09-15 | 0 | 0 | 0 |
| 2026-09-16 | 0 | 0 | 0 |
| 2026-09-17 | 0 | 0 | 0 |
| 2026-09-18 | 0 | 0 | 0 |
| 2026-09-19 | 0 | 0 | 1 |
| 2026-09-20 | 0 | 0 | 0 |
| 2026-09-21 | 0 | 0 | 0 |
| 2026-09-22 | 0 | 0 | 0 |
| 2026-09-23 | 0 | 0 | 0 |
| 2026-09-24 | 0 | 0 | 0 |
| 2026-09-25 | 0 | 0 | 0 |
| 2026-09-26 | 0 | 0 | 0 |
| 2026-09-27 | 0 | 0 | 0 |
| 2026-09-28 | 0 | 0 | 0 |
| 2026-09-29 | 0 | 0 | 0 |
| 2026-09-30 | 32K | 1 | 2 |
| 2026-10-01 | 0 | 0 | 0 |
| 2026-10-02 | 30K | 1 | 1 |

## Top countries: past 2 months

| Rank | Country | Tokens | Share |
| --- | --- | --- | --- |
| 1 | China | 0 | 15.1% |
| 2 | Indonesia | 0 | 13.4% |
| 3 | Taiwan | 0 | 12% |
| 4 | United States | 0 | 11% |
| 5 | India | 0 | 7.9% |
| 6 | Peru | 0 | 7.1% |
| 7 | Colombia | 0 | 4.2% |
| 8 | Egypt | 0 | 4% |
| 9 | Yemen | 0 | 3.1% |
| 10 | Russia | 0 | 2.6% |
| 11 | Brazil | 0 | 2.5% |
| 12 | Vietnam | 0 | 2.5% |
| 13 | Venezuela | 0 | 2.3% |
| 14 | Dominican Republic | 0 | 1.9% |
| 15 | Singapore | 0 | 1.8% |

## Nearby models by tokens last week

| Rank | Model | Lab | Tokens |
| --- | --- | --- | --- |
| 42 | [minimax-m2.5](https://opencode.ai/data/minimax/minimax-m2-5.md) | MiniMax | 720M |
| 43 | [gpt-5-nano](https://opencode.ai/data/openai/gpt-5-nano.md) | OpenAI | 289M |
| 44 | [test](https://opencode.ai/data/openai/test.md) | OpenAI | 218M |
| 45 | [exo](https://opencode.ai/data/unknown/exo.md) | - | 533K |
| 46 | [qwen3.8-27b](https://opencode.ai/data/alibaba/qwen3-8-27b.md) | Qwen | 62K |
| 47 | [test-novita-kimi k3](https://opencode.ai/data/moonshot/test-novita-kimi-k3.md) | Moonshot | 24K |
| 48 | [deepseek/deepseek-v4-flash](https://opencode.ai/data/deepseek/deepseek-deepseek-v4-flash.md) | DeepSeek | 16K |
| 49 | [tencent/hy3](https://opencode.ai/data/tencent/tencent-hy3.md) | Tencent | 16K |
| 50 | [ocz/muse-spark-1.3-contributor](https://opencode.ai/data/meta/ocz-muse-spark-1-3-contributor.md) | Meta | 12K |
| 51 | [test-novita-openai](https://opencode.ai/data/openai/test-novita-openai.md) | OpenAI | 451 |

## Benchmarks

| Benchmark | Score | Metric | Source |
| --- | --- | --- | --- |
| SWE-bench Pro | 61.7 | resolved | https://huggingface.co/Qwen/Qwen3.8-27B |
| Terminal-Bench | 73 | score | https://huggingface.co/Qwen/Qwen3.8-27B |
| NL2Repo | 42.3 | score | https://huggingface.co/Qwen/Qwen3.8-27B |
| DeepSWE | 42.2 | score | https://huggingface.co/Qwen/Qwen3.8-27B |
| QwenSWEBench | 79 | avg@3 | https://huggingface.co/Qwen/Qwen3.8-27B |
| CoWorkBench | 70.7 | score | https://huggingface.co/Qwen/Qwen3.8-27B |
| JobBench | 33.4 | score | https://huggingface.co/Qwen/Qwen3.8-27B |
| Agents' Last Exam | 20.4 | pass@1 | https://huggingface.co/Qwen/Qwen3.8-27B |
| Agents' Last Exam | 42.9 | score | https://huggingface.co/Qwen/Qwen3.8-27B |
| IFBench | 79.5 | score | https://huggingface.co/Qwen/Qwen3.8-27B |
| GPQA Diamond | 89.2 | score | https://huggingface.co/Qwen/Qwen3.8-27B |
| Humanity's Last Exam | 30.8 | score | https://huggingface.co/Qwen/Qwen3.8-27B |
| LiveCodeBench | 90.3 | score | https://huggingface.co/Qwen/Qwen3.8-27B |
| OSWorld-Verified | 84.3 | score | https://huggingface.co/Qwen/Qwen3.8-27B |
| WebArena-Verified | 64.8 | score | https://huggingface.co/Qwen/Qwen3.8-27B |
| AndroidWorld | 81.9 | score | https://huggingface.co/Qwen/Qwen3.8-27B |
| SWE-MM | 38.6 | score | https://huggingface.co/Qwen/Qwen3.8-27B |
| RealWorldQA | 85.9 | score | https://huggingface.co/Qwen/Qwen3.8-27B |
| ERQA | 65.5 | score | https://huggingface.co/Qwen/Qwen3.8-27B |
| RecreationBench | 47.1 | score | https://huggingface.co/Qwen/Qwen3.8-27B |
| ClawEval-MM | 57.4 | pass@3 | https://huggingface.co/Qwen/Qwen3.8-27B |
| ClawEval-MM | 56.9 | average score | https://huggingface.co/Qwen/Qwen3.8-27B |
| Vision2Web | 62.9 | score | https://huggingface.co/Qwen/Qwen3.8-27B |
| MathVision | 90 | score | https://huggingface.co/Qwen/Qwen3.8-27B |
| MathVision | 94.6 | score | https://huggingface.co/Qwen/Qwen3.8-27B |
| BabyVision | 65.7 | score | https://huggingface.co/Qwen/Qwen3.8-27B |
| BabyVision | 85.6 | score | https://huggingface.co/Qwen/Qwen3.8-27B |
| CharXiv Reasoning | 83.7 | score | https://huggingface.co/Qwen/Qwen3.8-27B |
| CharXiv Reasoning | 90.2 | score | https://huggingface.co/Qwen/Qwen3.8-27B |
| OmniDocBench | 91.1 | score | https://huggingface.co/Qwen/Qwen3.8-27B |

## Methodology

- Updates: Aggregated every hour. Days and weeks use UTC.
- Tokens: Input, output, reasoning, and cached tokens for each request.
- Users and sessions: Approximate counts of distinct users and OpenCode sessions.
- Cost: Session cost is the average cost per OpenCode session. Token prices are list prices from the OpenCode model catalog.
- Retention: The share of a model's users in one week who use it again the next week.
- Citation: Cite OpenCode Data (opencode.ai/data) with the update time shown at the top of the page.