NOUMENOpen app
NOUMEN

Qwen · Open weights

Qwen3.6 35B A3B Thinking

Added Apr 19, 2026Open weights

Compare models

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

Loading models…

Use in the app

Qwen3.6 35B A3B is a native vision-language MoE model with hybrid attention. Compared to Qwen3.5 35B A3B, Alibaba reports stronger agentic coding, mathematical and code reasoning, and better spatial understanding (including object localization and detection).

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

Pricing

USD per 1M tokens
Input
$0.11
Output
$0.80

Reference prices as published by the model provider. API example

Specifications

Context window
1M
Max output
16.4K

Side by side

Qwen3.6 35B A3B Thinking 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…

Qwen3.6 35B A3B ThinkingQwenThis modelMistral Small 3 24B (2501)MistralRemove
ProviderQwenMistral
Discovery tierOpen weightsOpen weights
AddedApr 19, 2026Sep 10, 2026
Limits
Context window1.05MBest32.77K
Max output16.38KBest8.19K
Average output—403 tokens
Pricing, USD per 1M tokens
Input$0.11Best$0.12
Output$0.80$0.35Best
Performance
Request success (24h)100.0%—
Benchmark indexes
Intelligence Index—6.7
Math Index—4.3
Capabilities
ReasoningYesNo
ToolsYesYes
VisionYesNo
Video inputYesNo
Reasoning effort——

Public evaluations

Benchmarks

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

  • Qwen3.6 35B A3B Thinking
  • Mistral Small 3 24B (2501)

Intelligence Index

  • Qwen3.6 35B A3B Thinking—
  • Mistral Small 3 24B (2501)6.7
3.8Catalog models by score55.1

Indexes

  • Math Index

    Composite of the math evaluations

    • Qwen3.6 35B A3B Thinking: —
    • Mistral Small 3 24B (2501): 4.3

Document reasoning

  • AA-LCRv1.1

    Reasoning across long documents

    • Qwen3.6 35B A3B Thinking: —
    • Mistral Small 3 24B (2501): 0.0%

Reasoning and knowledge

  • Humanity’s Last Exam

    Expert-level questions across academic fields

    • Qwen3.6 35B A3B Thinking: —
    • Mistral Small 3 24B (2501): 3.8%

  • GPQA Diamond

    Graduate-level science questions

    • Qwen3.6 35B A3B Thinking: —
    • Mistral Small 3 24B (2501): 46.2%

  • MMLU-Pro

    Multi-subject knowledge and reasoning

    • Qwen3.6 35B A3B Thinking: —
    • Mistral Small 3 24B (2501): 65.2%

  • IFBench

    Following precise output instructions

    • Qwen3.6 35B A3B Thinking: —
    • Mistral Small 3 24B (2501): 26.4%

Coding

  • LiveCodeBench

    Recent competitive programming problems

    • Qwen3.6 35B A3B Thinking: —
    • Mistral Small 3 24B (2501): 25.2%

Math

  • AIME 2025

    Competition mathematics, 2025 problems

    • Qwen3.6 35B A3B Thinking: —
    • Mistral Small 3 24B (2501): 4.3%

  • AIME

    Competition mathematics

    • Qwen3.6 35B A3B Thinking: —
    • Mistral Small 3 24B (2501): 8.0%

  • MATH-500

    Competition math problems

    • Qwen3.6 35B A3B Thinking: —
    • Mistral Small 3 24B (2501): 71.5%

Legacy evaluations

  • τ²-Bench Telecom

    Tool use in a customer-support simulation

    • Qwen3.6 35B A3B Thinking: —
    • Mistral Small 3 24B (2501): 19.6%

  • AA-LCR (unversioned)

    Long-context reasoning, earlier grading

    • Qwen3.6 35B A3B Thinking: —
    • Mistral Small 3 24B (2501): 0.0%

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.

Prompt cache hits
66.6%
Request success (24h)
100.0%
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 100.0%
  • 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/qwen3.6-35b-a3b:thinking",
    "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/qwen3.6-35b-a3b:thinking",
    messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)

Qwen

Other Qwen models, starting with the same family.