Qwen · Efficient
Qwen3.5 Flash Thinking
Qwen3.5 Flash with extended reasoning enabled. The fastest and cheapest native Qwen 3.5 vision-language model.
- Reasoning
- Vision
- Video input
- Input
- Text, Images, Video
- Output
- Text
- Discovery tier
- Efficient
Pricing
USD per 1M tokens- Input
- $0.10
- Output
- $0.40
Reference prices as published by the model provider. API example
Specifications
- Context window
- 991.8K
- Max output
- 65.5K
Side by side
Qwen3.5 Flash Thinking compared
Specifications, prices, speed and benchmark scores next to each other. Benchmarks below show every compared model too.
| Qwen3.5 Flash ThinkingQwenThis model | GPT 6 LunaOpenAIRemove | |
|---|---|---|
| Provider | Qwen | OpenAI |
| Discovery tier | Efficient | Efficient |
| Added | Feb 24, 2026 | Sep 22, 2026 |
| Limits | ||
| Context window | 991.81K | 1.05MBest |
| Max output | 65.54K | 128KBest |
| Average output | — | 120 tokens |
| Pricing, USD per 1M tokens | ||
| Input | $0.10 | $0.10 |
| Output | $0.40Best | $0.50 |
| Cache read | — | $0.01 |
| Cache write | — | $0.13 |
| Performance | ||
| Throughput | — | 296.3 tok/s |
| Time to first token | — | 3.8 s |
| Request success (24h) | — | 99.7% |
| Benchmark indexes | ||
| Intelligence Index | — | 29.9 |
| Capabilities | ||
| Reasoning | Yes | Yes |
| Tools | No | Yes |
| Vision | Yes | Yes |
| Structured output | No | Yes |
| PDF input | No | Yes |
| Video input | Yes | No |
| Reasoning effort | — | none, low, medium, high, xhigh, max |
Public evaluations
Benchmarks
Independent evaluation results, with how each score ranks among the models in the Noumen catalog.
- Qwen3.5 Flash Thinking
- GPT 6 Luna
Intelligence Index
- Qwen3.5 Flash Thinking—
- GPT 6 Luna29.9
Agentic work
AutomationBench-AA
Workflow automation, penalized for guardrail violations
- Qwen3.5 Flash Thinking: —
- GPT 6 Luna: 40.5%
AutomationBench-AA tasks completed
Workflows finished end to end without a guardrail violation
- Qwen3.5 Flash Thinking: —
- GPT 6 Luna: 10.2%
AA-Briefcasev1.1
Agentic knowledge work, rated head to head
- Qwen3.5 Flash Thinking: —
- GPT 6 Luna: 1,097 Elo
GDPval-AAv2.1
Economically valuable occupational tasks, rated head to head
- Qwen3.5 Flash Thinking: —
- GPT 6 Luna: 1,263 Elo
Document reasoning
GDP.pdf
Professional reasoning over PDF documents, all-pass rate
- Qwen3.5 Flash Thinking: —
- GPT 6 Luna: 13.8%
AA-LCRv1.1
Reasoning across long documents
- Qwen3.5 Flash Thinking: —
- GPT 6 Luna: 78.3%
MLCR-AA
Long-context reasoning over medical records
- Qwen3.5 Flash Thinking: —
- GPT 6 Luna: 18.3%
Reasoning and knowledge
Humanity’s Last Exam
Expert-level questions across academic fields
- Qwen3.5 Flash Thinking: —
- GPT 6 Luna: 28.3%
Coding
Terminal-Benchv4.0
Practical software engineering in a terminal
- Qwen3.5 Flash Thinking: —
- GPT 6 Luna: 2.5%
SciCode
Scientific computing in Python
- Qwen3.5 Flash Thinking: —
- GPT 6 Luna: 50.9%
Legacy evaluations
AA-LCR (unversioned)
Long-context reasoning, earlier grading
- Qwen3.5 Flash Thinking: —
- GPT 6 Luna: 78.3%
Scores from Artificial Analysis, updated Oct 26, 2026. Artificial Analysis
- Qwen3.5 Flash Thinking
- GPT 6 Luna
| Leaderboard | Qwen3.5 Flash Thinking | GPT 6 Luna |
|---|---|---|
| ARC-AGI-3 (Default harness) | Not in top 20 | #18 · 0.2% |
Ranks among Noumen catalog models on ARC Prize leaderboards. Only top-20 placements are listed.ARC Prize
- Qwen3.5 Flash Thinking
- GPT 6 Luna
| Leaderboard | Qwen3.5 Flash Thinking | GPT 6 Luna |
|---|---|---|
| SVG | Not in top 20 | #16 · 1250 |
Ranks among Noumen catalog models on Design Arena leaderboards. Only top-20 placements are listed.Design Arena
Developers
Use it through the API
Noumen’s API is OpenAI-compatible: any Chat Completions client works with a base URL and a key.
curl https://www.noumen.xyz/api/v1/chat/completions \
-H "Authorization: Bearer $NOUMEN_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "qwen/qwen3.5-flash: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/qwen3.5-flash:thinking",
messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)Qwen
Related models
Other Qwen models, starting with the same family.