NOUMENOpen app
NOUMEN

Qwen · Open weights

Qwen 3.8 27B Swift

Added Oct 5, 2026Open weights

Compare models

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

Loading models…

Use in the app

Qwen 3.8 27B Swift is a finetune for fast responses, served by Aoru in NVFP4 with a 262,144-token context window. It supports image understanding, tool calling, and optional reasoning.

  • Reasoning
  • Tools
  • Vision
Input
Text, Images
Output
Text
Discovery tier
Open weights

Pricing

USD per 1M tokens
Input
$0.10
Output
$0.60
Cache read
$0.05

Reference prices as published by the model provider. API example

Specifications

Context window
262.1K
Max output
32.8K
Average output (7 days)
158 tokens
Longer than 15% of 468 models

Side by side

Qwen 3.8 27B Swift 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…

Qwen 3.8 27B SwiftQwenThis modelGLM 5.3 FlashZ.aiRemove
ProviderQwenZ.ai
Discovery tierOpen weightsOpen weights
AddedOct 5, 2026Aug 26, 2026
Limits
Context window262.14K1.05MBest
Max output32.77K131.07KBest
Average output158 tokens1.3K tokens
Pricing, USD per 1M tokens
Input$0.10$0.10
Output$0.60$0.30Best
Cache read$0.05$0.025Best
Performance
Throughput41.8 tok/s104.6 tok/sBest
Time to first token1.7 sBest4.0 s
Request success (24h)90.9%99.6%Best
Benchmark indexes
Intelligence Index—41.8
Coding Index—71.5
Capabilities
ReasoningYesYes
ToolsYesYes
VisionYesYes
Structured outputNoYes
Video inputNoYes
Reasoning effort—low, high, max

Public evaluations

Benchmarks

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

  • Qwen 3.8 27B Swift
  • GLM 5.3 Flash

Intelligence Index

  • Qwen 3.8 27B Swift—
  • GLM 5.3 Flash41.8
3.8Catalog models by score55.1

Indexes

  • Coding Index

    Composite of the coding evaluations

    • Qwen 3.8 27B Swift: —
    • GLM 5.3 Flash: 71.5

Agentic work

  • AutomationBench-AA

    Workflow automation, penalized for guardrail violations

    • Qwen 3.8 27B Swift: —
    • GLM 5.3 Flash: 60.4%

  • AutomationBench-AA tasks completed

    Workflows finished end to end without a guardrail violation

    • Qwen 3.8 27B Swift: —
    • GLM 5.3 Flash: 25.4%

  • Harvey LAB-AAv1.1

    Legal tasks passed in full, with no material hallucination

    • Qwen 3.8 27B Swift: —
    • GLM 5.3 Flash: 0.8%

  • AA-Briefcasev1.1

    Agentic knowledge work, rated head to head

    • Qwen 3.8 27B Swift: —
    • GLM 5.3 Flash: 1,452 Elo

  • GDPval-AAv2.1

    Economically valuable occupational tasks, rated head to head

    • Qwen 3.8 27B Swift: —
    • GLM 5.3 Flash: 1,642 Elo

Document reasoning

  • GDP.pdf

    Professional reasoning over PDF documents, all-pass rate

    • Qwen 3.8 27B Swift: —
    • GLM 5.3 Flash: 15.4%

  • AA-LCRv1.1

    Reasoning across long documents

    • Qwen 3.8 27B Swift: —
    • GLM 5.3 Flash: 80.0%

  • MLCR-AA

    Long-context reasoning over medical records

    • Qwen 3.8 27B Swift: —
    • GLM 5.3 Flash: 11.7%

Reasoning and knowledge

  • Humanity’s Last Exam

    Expert-level questions across academic fields

    • Qwen 3.8 27B Swift: —
    • GLM 5.3 Flash: 39.9%

  • GPQA Diamond

    Graduate-level science questions

    • Qwen 3.8 27B Swift: —
    • GLM 5.3 Flash: 91.2%

Coding

  • Terminal-Benchv4.0

    Practical software engineering in a terminal

    • Qwen 3.8 27B Swift: —
    • GLM 5.3 Flash: 32.8%

  • SciCode

    Scientific computing in Python

    • Qwen 3.8 27B Swift: —
    • GLM 5.3 Flash: 51.6%

Legacy evaluations

  • AA-LCR (unversioned)

    Long-context reasoning, earlier grading

    • Qwen 3.8 27B Swift: —
    • GLM 5.3 Flash: 80.0%

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.

Throughput
41.8 tok/s
Faster than 30% of 254 models
Time to first token
1.7 s
Quicker than 86% of 254 models
Prompt cache hits
71.0%
About $0.0645 per 1M input tokens with caching
Request success (24h)
88.4%
Unstable

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 33.3%
  • 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-swift",
    "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-swift",
    messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)

Qwen

Other Qwen models, starting with the same family.