qwen/qwen3.8-max-0902

Model release OpenRouter 模型 Source language: Chinese

OpenRouter 于 2026-09-04 收录该模型(收录时间不等于模型正式发布日期);上下文:1,000,000 tokens;输入价:$2.0000/百万 tokens;输出价:$6.0000/百万 tokens;定位(来源描述):Qwen3.8 Max 0902 is an updated snapshot of Qwen3.8 Max from Alibaba's Qwen team.

Full text

Qwen: Qwen3.8 Max (0902)

qwen/qwen3.8-max-0902

Compare

Qwen3.8 Max 0902 is an updated snapshot of Qwen3.8 Max from Alibaba's Qwen team. It is a 2.4-trillion-parameter mixture-of-experts model that accepts text, image, and video input and returns text, with a 1M-token context window and reasoning enabled by default.

This snapshot is post-trained for coding and agentic work, including multi-step software projects, multi-tool orchestration, and long-horizon task execution. It also targets chart reasoning, document parsing, and multimodal understanding over long documents and extended video. Tool calling, structured outputs, and configurable reasoning effort are supported.

Modalities

In / out price

$2 / $6per 1M

Context

1.0M

Released

Sep 3, 2026

Compare

Providers

Different companies host the same model. OpenRouter routes your request to one of them based on the routing mode you pick: Balanced (price + speed), Nitro (fastest), or Floor (cheapest).

Latency / throughput

$2.00$6.00$0.25$2.50$0.171.75s38 tok/s

99.07%

$2.00$6.00$0.25----1.99s32 tok/s

97.77%

ProviderInput / Output /M

Input $2.00 / Output $6.00

38 tok/s99.07% uptime

Input $2.00 / Output $6.00

32 tok/s97.77% uptime

Pricing

The average price customers actually pay for this model, next to the prices providers post. Caching and discounts mean the price actually paid is often well below the listed one.

Weighted average

Input

$0.4414/M tokens

Output

$6/M tokens

Qwen3.8 Max (0902) — Price history

Price history

No effective price data for this selection

$0.440$6.0089.2%80.3%

$0.449$6.0088.4%19.7%

ProviderInput / Output /M

Token share

Effective in $0.440 / Effective out $6.00

Token share

80.3%

Effective in $0.449 / Effective out $6.00

Token share

19.7%

Performance

Throughput is how fast the model writes (tokens per second — higher is better). Latency is total round-trip time (lower is better). TTFT is time-to-first-token — how long before you see anything appear (lower is better).

Throughput

38tok/s

P50, best across providers

Latency

1.75s

P50, best provider

Reasoning effortLatency / throughput

Uptime

Uptime is the percentage of the past 3 days that at least one provider was responding to requests. Availability is the percentage of time that inference was successfully served. OpenRouter continuously monitors and uses the next-best provider when one returns an error.

Uptime (3d)The model was reachable. Request routed to a provider.

100.00%

Availability (3d)The model returned inference from any provider. Errors and empty responses count against it.

99.35%

Availability over the last 3 days

Last 72 hours

Availability 99.35%

3 Days Ago2 Days AgoYesterdayNow

Availability over the last 24 hours

OpenRouter Availability

99.49%

When an error occurs in an upstream provider, we can recover by routing to another healthy provider, if your request filters allow it. You can access per-provider uptime data programmatically through the Endpoints API. Learn more about our load balancing and provider routing options.

Benchmarks

Scores on standardized evaluations and this model’s own outputs on shared benchmark prompts. Higher percentages are better — and each rank shows where this model lands among the models evaluated on that source.

Benchmark score summary for Qwen: Qwen3.8 Max (0902) (Artificial Analysis and Vals AI)

SourceBenchmarkScore

Artificial AnalysisQwen3.8 Max (0902) Intelligence Index33.7

Artificial AnalysisQwen3.8 Max (0902) Coding Index76.2

Artificial AnalysisQwen3.8 Max (0902) Agentic Index56.0

Artificial AnalysisQwen3.8 Max (0902) GPQA Diamond92.8%

Artificial AnalysisQwen3.8 Max (0902) HLE43.1%

Artificial AnalysisQwen3.8 Max (0902) AA-LCR80.3%

Artificial AnalysisQwen3.8 Max (0902) τ-Bench Banking47.8%

Artificial AnalysisQwen3.8 Max (0902) GDPval-AA58.7%

Artificial AnalysisQwen3.8 Max (0902) CritPt17.7%

Artificial AnalysisQwen3.8 Max (0902) SciCode52.1%

Artificial AnalysisQwen3.8 Max (0902) Terminal-Bench 4.038.9%

Artificial AnalysisQwen3.8 Max (0902) Terminal-Bench 2.188.8%

Artificial AnalysisQwen3.8 Max (0902) AA-Omniscience Accuracy29.5%

Artificial AnalysisQwen3.8 Max (0902) AA-Omniscience Non-Hallucination Rate72.9%

Vals AIQwen 3.8 Max Access to Justice Benchmark76.4%

Vals AIQwen 3.8 Max Code Migration24.0%

Vals AIQwen 3.8 Max CorpFin v265.9%

Vals AIQwen 3.8 Max CyberBench v1.128.6%

Vals AIQwen 3.8 Max EMB v1.166.0%

Vals AIQwen 3.8 Max Finance Agent (v2)50.6%

Vals AIQwen 3.8 Max Financial Coaching Bench75.6%

Vals AIQwen 3.8 Max GPQA Diamond93.7%

Vals AIQwen 3.8 Max Harvey's Legal Agent Benchmark v110.4%

Vals AIQwen 3.8 Max Immigration70.1%

Vals AIQwen 3.8 Max IOI68.9%

Vals AIQwen 3.8 Max LiveCodeBench87.9%

Vals AIQwen 3.8 Max LegalBench83.6%

Vals AIQwen 3.8 Max Legal Research Bench47.6%

Vals AIQwen 3.8 Max MedCode40.7%

Vals AIQwen 3.8 Max MedScribe84.9%

Vals AIQwen 3.8 Max MMLU Pro88.6%

Vals AIQwen 3.8 Max MMMU Pro88.0%

Vals AIQwen 3.8 Max MortgageTax64.0%

Vals AIQwen 3.8 Max MysteryMechanism23.9%

Vals AIQwen 3.8 Max ProgramBench0.0%

Vals AIQwen 3.8 Max ProofBench v1.158.0%

Vals AIQwen 3.8 Max Public Benefits Bench v1.167.1%

Vals AIQwen 3.8 Max Vals RSI Index21.8%

Vals AIQwen 3.8 Max SAGE81.8%

Vals AIQwen 3.8 Max SkillsBench42.0%

Vals AIQwen 3.8 Max SWE-bench85.6%

Vals AIQwen 3.8 Max SWE-bench Multimodal42.5%

Vals AIQwen 3.8 Max Tax Agent Bench32.1%

Vals AIQwen 3.8 Max TaxEval v275.6%

Vals AIQwen 3.8 Max Terminal-Bench 4.034.3%

Vals AIQwen 3.8 Max Terminal-Bench-Science12.9%

Vals AIQwen 3.8 Max Vals Index48.3%

Vals AIQwen 3.8 Max Vals Multimodal Index65.4%

Vals AIQwen 3.8 Max Vibe Code Bench 1-10012.8%

Vals AIQwen 3.8 Max Vibe Code Bench v1.164.7%

Apps

Public apps that send the most traffic to this model. Good signal for what real production workloads look like — and a hint at which use cases this model is best suited for.

1.

Hermes Agent

Hermes Agent is an open-source, self-improving AI agent by Nous Research that runs persistently with memory across sessions, and builds reusable skills from experience. It comes with 40+ built-in tools, including web search, browser automation, and vision, plus scheduled automations and subagents.

96.2Btokens

2.

Claude Code

Claude Code is Anthropic's agentic coding tool that reads your entire codebase, plans and executes changes across files, runs tests, and iterates on failures, all from natural language prompts.

29Btokens

3.

pi

There are many coding agents, but this one is yours.

11.2Btokens

4.

OpenClaw

OpenClaw is an open-source AI agent that connects to your messaging apps and takes real actions on your behalf, from running commands and browsing the web to managing files and sending emails.

8.29Btokens

5.

DeepSeek Harness

new

5.48Btokens

Rest of OpenRouter

1.48B

Top 5 Apps

668M

Activity

Token volume and request traffic to this model over time.

API

Drop-in code to call this model. OpenRouter's API is OpenAI-compatible — most SDKs work by just swapping the base URL. The only thing that changes between models is the model slug below.

Not currently available through EU in-region routing. In-region routing docs

Frequently asked questions

Qwen3.8 Max 0902 is an updated snapshot of Qwen3.8 Max from Alibaba's Qwen team. It is a 2.4-trillion-parameter mixture-of-experts model that accepts text, image, and video input and returns text, with a 1M-token context window and reasoning enabled by default.

Qwen3.8 Max (0902) costs $2.00/M input tokens and $6.00/M output tokens, with separate rates for Cache Read at $0.25/M tokens.

Qwen3.8 Max (0902) has a 1,000,000 token context window. It supports up to 131,072 completion tokens.

Yes. Qwen3.8 Max (0902) accepts tools and tool_choice for function calling. It also supports structured outputs via a JSON schema in response_format.

Qwen3.8 Max (0902) accepts text, images, and video as input and returns text.

Qwen3.8 Max (0902) is served by 1 provider on OpenRouter: Alibaba Cloud Int..

Qwen4 Max, Qwen3.8 Max Prime, Qwen3.8 Omni Flash and 36 more are other text models from Qwen.

Qwen3.8 Max (0902) was released on September 3, 2026.

Sources

← Back to the daily brief