Qwen3.8 Flash

Name: Qwen3.8 Flash
Author: Alibaba

Multimodal

Alibaba

Qwen3.8 Flash is the production QwenCloud / OpenRouter API version of Qwen's 125B-parameter MoE Flash architecture, built on the open-weight Qwen3.8-Flash-Next checkpoint. It ships with a 1M-token default context, built-in tools, function calling, and structured output, with thinking enabled by default. Multimodal text, image, and video understanding at roughly one-ninth the training cost of Qwen3.7-Plus.

Key Specifications

Parameters

125.0B

Context

1.0M

Release Date

August 26, 2026

Average Score

68.5%

API Documentation Model Weights Results Blog

Timeline

Key dates in the model's history

Announcement

August 26, 2026

Last Update

August 31, 2026

Technical Specifications

Parameters

125.0B

Training Tokens

Knowledge Cutoff

Family

Capabilities

MultimodalZeroEval

Pricing & Availability

Input (per 1M tokens)

$0.15

Output (per 1M tokens)

$0.47

Max Input Tokens

1.0M

Max Output Tokens

131.1K

Supported Features

Function CallingStructured OutputCode ExecutionWeb SearchBatch InferenceFine-tuning

Benchmark Results

Model performance metrics across various tests and benchmarks

Reasoning

Logical reasoning and analysis

GPQA

Diamond split • Self-reported

91.7%

Other Tests

Specialized benchmarks

Agents' Last Exam

Score • Self-reported

51.2%

AndroidWorld

• Self-reported

84.5%

CharXiv-R

With code interpreter • Self-reported

90.6%

ClawEval-MM

Average score across three trials • Self-reported

60.4%

CoWorkBench

• Self-reported

73.9%

DeepSWE 1.1

Best of Claude Code and mini-SWE-agent, temp=1.0, top_p=0.95, 256K context • Self-reported

58.7%

ERQA

• Self-reported

72.3%

Humanity's Last Exam

Judged by GPT-4o • Self-reported

35.9%

IFBench

• Self-reported

81.3%

Job Bench

• Self-reported

55.7%

LiveCodeBench v6

• Self-reported

91.9%

LVBench

• Self-reported

76.6%

MathVision

With code interpreter • Self-reported

95.7%

NL2Repo

Claude Code harness • Self-reported

48.1%

OSWorld 2.0

Binary completion • Self-reported

19.4%

RealWorldQA

• Self-reported

88.5%

RecreationBench

• Self-reported

49.9%

SWE-bench Multilingual

mini-SWE-agent, temp=1.0, top_p=0.95, 256K context • Self-reported

81.0%

SWE-Bench Pro

Claude Code, temp=1.0, top_p=0.95, 256K context • Self-reported

62.5%

Toolathlon

Verified split; Pass@1 • Self-reported

73.5%

Vision2Web

Claude Code harness, judged by gpt-5.4-2026-03-05 • Self-reported

64.0%

License & Metadata

License

proprietary

Announcement Date

August 26, 2026

Last Updated

August 31, 2026

Compare Qwen3.8 Flash

All comparisons

vs Qwen3.8 Flash Next vs Qwen3.8-27B vs Qwen3.8 Max vs Qwen3.5 35B A3B vs Qwen3.5 122B A10B vs Qwen3.5 27B vs Qwen3.7 Max vs Qwen3.6-27B

Similar Models

All Models

Qwen3.8 Flash Next

Alibaba

MM180.0B

Best score:0.9 (GPQA)

Released:Aug 2026

Price:$0.16/1M tokens

Qwen3.5-397B-A17B

Alibaba

MM397.0B

Released:Feb 2026

Qwen3.8 Max

Alibaba

MM2.4T

Released:Aug 2026

Price:$2.50/1M tokens

Qwen3.7-Plus

Alibaba

Best score:0.9 (GPQA)

Released:May 2026

Price:$0.32/1M tokens

Qwen3.6-35B-A3B

Alibaba

MM35.0B

Best score:0.9 (GPQA)

Released:Apr 2026

Qwen3.8-27B

Alibaba

MM27.8B

Best score:0.9 (GPQA)

Released:Aug 2026

Qwen3 235B A22B

Alibaba

235.0B

Best score:0.9 (MMLU)

Released:Apr 2025

Price:$0.20/1M tokens

Inkling-Small

Thinking Machines Lab

MM276.0B

Best score:0.9 (GPQA)

Released:Jul 2026

Price:$0.30/1M tokens

Recommendations are based on similarity of characteristics: developer organization, multimodality, parameter size, and benchmark performance. Choose a model to compare or go to the full catalog to browse all available AI models.