NOUMENOpen app
NOUMEN

Qwen · Open weights

Qwen3.8 27B

Added Aug 16, 2026Open weights

Compare models

Choose up to 3 models to compare.1 of 3 chosen

Loading models…

Use in the app

Qwen3.8 27B is an open-weight multimodal model from Alibaba for coding, visual understanding, tool use, and structured output. This variant keeps thinking disabled for faster direct responses. Requests containing video cost $0.32 per million input tokens and $2.50 per million output tokens.

  • Reasoning
  • Tools
  • Vision
  • Structured output
  • Video input
Input
Text, Images, Video
Output
Text
Discovery tier
Open weights

Pricing

USD per 1M tokens
Input
$0.15
Output
$0.70
Cache read
$0.04

Reference prices as published by the model provider. API example

Specifications

Context window
1M
Max output
32.8K

Side by side

Qwen3.8 27B compared

Specifications, prices, speed and benchmark scores next to each other. Benchmarks below show every compared model too.

Compare models

Choose up to 3 models to compare.1 of 3 chosen

Loading models…

Qwen3.8 27BQwenThis modelGemma 4 31B IT TEEGoogleRemove
ProviderQwenGoogle
Discovery tierOpen weightsOpen weights
AddedAug 16, 2026May 26, 2026
Limits
Context window1.05MBest262.14K
Max output32.77K262.14KBest
Average output—269 tokens
Pricing, USD per 1M tokens
Input$0.15$0.15
Output$0.70$0.46Best
Cache read$0.04—
Performance
Throughput—56.1 tok/s
Time to first token—6.0 s
Request success (24h)99.9%Best99.8%
Benchmark indexes
Intelligence Index—14.7
Coding Index—43.4
Capabilities
ReasoningYesYes
ToolsYesYes
VisionYesNo
Structured outputYesNo
Video inputYesNo
Reasoning effort——

Public evaluations

Benchmarks

Independent evaluation results, with how each score ranks among the models in the Noumen catalog.

  • Qwen3.8 27B
  • Gemma 4 31B IT TEE

Intelligence Index

  • Qwen3.8 27B—
  • Gemma 4 31B IT TEE14.7
3.8Catalog models by score55.1

Indexes

  • Coding Index

    Composite of the coding evaluations

    • Qwen3.8 27B: —
    • Gemma 4 31B IT TEE: 43.4

Agentic work

  • AutomationBench-AA

    Workflow automation, penalized for guardrail violations

    • Qwen3.8 27B: —
    • Gemma 4 31B IT TEE: 4.9%

  • AutomationBench-AA tasks completed

    Workflows finished end to end without a guardrail violation

    • Qwen3.8 27B: —
    • Gemma 4 31B IT TEE: 0.5%

  • AA-Briefcasev1.1

    Agentic knowledge work, rated head to head

    • Qwen3.8 27B: —
    • Gemma 4 31B IT TEE: 368 Elo

  • GDPval-AAv2.1

    Economically valuable occupational tasks, rated head to head

    • Qwen3.8 27B: —
    • Gemma 4 31B IT TEE: 619 Elo

Document reasoning

  • GDP.pdf

    Professional reasoning over PDF documents, all-pass rate

    • Qwen3.8 27B: —
    • Gemma 4 31B IT TEE: 6.0%

  • AA-LCRv1.1

    Reasoning across long documents

    • Qwen3.8 27B: —
    • Gemma 4 31B IT TEE: 69.7%

Reasoning and knowledge

  • Humanity’s Last Exam

    Expert-level questions across academic fields

    • Qwen3.8 27B: —
    • Gemma 4 31B IT TEE: 23.6%

  • GPQA Diamond

    Graduate-level science questions

    • Qwen3.8 27B: —
    • Gemma 4 31B IT TEE: 85.7%

  • IFBench

    Following precise output instructions

    • Qwen3.8 27B: —
    • Gemma 4 31B IT TEE: 75.6%

Coding

  • Terminal-Benchv4.0

    Practical software engineering in a terminal

    • Qwen3.8 27B: —
    • Gemma 4 31B IT TEE: 0.0%

  • SciCode

    Scientific computing in Python

    • Qwen3.8 27B: —
    • Gemma 4 31B IT TEE: 45.5%

Legacy evaluations

  • Terminal-Bench Hard

    Earlier terminal task suite

    • Qwen3.8 27B: —
    • Gemma 4 31B IT TEE: 36.4%

  • τ²-Bench Telecom

    Tool use in a customer-support simulation

    • Qwen3.8 27B: —
    • Gemma 4 31B IT TEE: 59.9%

  • AA-LCR (unversioned)

    Long-context reasoning, earlier grading

    • Qwen3.8 27B: —
    • Gemma 4 31B IT TEE: 69.7%

Scores from Artificial Analysis, updated Oct 26, 2026. Artificial Analysis

Live routing data

Performance and reliability

Recent averages across requests routed for this model. Your own speed depends on prompt length, reasoning effort and load.

Prompt cache hits
40.5%
About $0.11 per 1M input tokens with caching
Request success (24h)
99.9%
Reliable

Cache hits are the share of input tokens recently served from a prompt cache across all requests. They are not a forecast of your own hit rate.

Hourly request success · last 72 hoursLowest hour 98.3%
  • Healthy (98% or more)
  • Degraded (90–98%)
  • Down (under 90%)
  • No requests

Developers

Use it through the API

Noumen’s API is OpenAI-compatible: any Chat Completions client works with a base URL and a key.

Read the quickstart
curl https://www.noumen.xyz/api/v1/chat/completions \
  -H "Authorization: Bearer $NOUMEN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "qwen/qwen3.8-27b",
    "messages": [{ "role": "user", "content": "Explain this stack trace." }]
  }'
import os
from openai import OpenAI

client = OpenAI(base_url="https://www.noumen.xyz/api/v1", api_key=os.environ["NOUMEN_API_KEY"])
reply = client.chat.completions.create(
    model="qwen/qwen3.8-27b",
    messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)

Qwen

Other Qwen models, starting with the same family.