Qwen3 VL 235B A22B (Reasoning)
Sep 2025
Older versionReplace withQwen3.5 397B A17BNewer version
MiniMax M3.1 FlashCheaper, as good
GPT-6 LunaBetter
MiMo v2.6 ProOpen weights
Qwen3.5 397B A17BNewer versionMiniMax M3.1 FlashCheaper, as goodGPT-6 LunaBetterMiMo v2.6 ProOpen weightsQwen3 VL 235B A22B (Reasoning)Compare inon
Benchmarks of Qwen3 VL 235B A22B (Reasoning)14
BenchmarkByScoreRank
Overall1
Knowledge4
Science2
Coding1
Multimodal1
Documents1
Legacy · saturated or no longer reported4
No benchmark matches
Ranking?
Retired models are left out of the current rankings.
Alternatives?
ModelScoreCostPer taskWhat changes
CheaperLing 3.1 Flash 1019×3.3-10%$1.49BetterGPT-6 Lunalow+60%-60%$0.005Function calling only with none · Chat Completions+3OpenMiMo v2.6 Prothinking×3.4+8%$0.13Pass back reasoning_content+3
CheaperClaude Haiku 5.5max×3.2-60%$0.21Remove sampling parameters+5CheaperMiniMax M3.1 Flashthinking×3.2-48%$0.22No strict JSON+1ModelScoreCostPer taskWhat changes
Ling 3.1 Flash 1019×3.3-10%$1.49Claude Haiku 5.5max×3.2-60%$0.21Remove sampling parameters+5MiniMax M3.1 Flashthinking×3.2-48%$0.22No strict JSON+1
GLM 5.3 FlashmaxOpen×3.1-53%$0.28Defaults to max · Self-hosted+4Ling 3.1 FlashOpen×3.1-10%$1.01ModelScoreCostPer taskWhat changes
GPT-6 Lunalow+60%-60%$0.005Function calling only with none · Chat Completions+3GPT-5.6 Lunalow+56%-10%$0.01Claude Haiku 5.5low×2.2-60%$0.024Remove sampling parameters+5Ling 3.0 FlashthinkingOpen+50%-82%$0.028MiMo v2.5thinkingOpen+62%-65%$0.042ModelScoreCostPer taskWhat changes
MiMo v2.6 ProthinkingOpen×3.4+8%$0.13Pass back reasoning_content+3
DeepSeek V4.1 Flash 1017maxOpen×3.4+5%$0.26Pass back reasoning_content · Chat Completions+4GLM 5.3maxOpen×3.3+329%$2.01Reasoning cannot be disabled+3
Step 5 PreviewthinkingOpen×3.3+184%$1.03web_search not available · Chat Completions+2
Kimi K3maxOpen×3.2+1,096%$2.00Sampling parameters are fixed+3ModelScoreCostPer taskWhat changes
Qwen3.8 Max (0902)thinking×3.4—$5.41Qwen3.8 2.4T A95BthinkingOpen×3—$2.16max_tokens range · QwenCloud+4Qwen3.8 FlashthinkingOpen×3—$0.37max_tokens range · QwenCloud+5Qwen3.7 Maxthinking×2.2—$0.93Qwen3.6 Max Previewthinking×2.1—$1.30 / $7.80Include the word JSON · QwenCloudQwen3.8 27BmediumOpen×2.1—$1.27max_tokens range · QwenCloud+4Qwen3.6 Plusthinking×2—$0.50 / $3Qwen3.7 Plusthinking+87%—$0.22All Alibaba (Qwen) models →Run it yourself
Fits onhardware needs not verified yet
Licencenot verified
Benchmarks
all categoriesChat & Knowledge
AA Omniscience Index-46.6
GPQA Diamond77.2
Coding
Agents & Tool Use
τ²-bench · AA54.1
Writing & Language
Vision & Documents
FAQ
How much does Qwen3 VL 235B A22B (Reasoning) cost?›
$0.29 per 1M input tokens and $1.15 per 1M output tokens (cheapest host: Alibaba (China)).
Is Qwen3 VL 235B A22B (Reasoning) free to use?›
There is no free API; the cheapest paid price is $0.29 per 1M input tokens. The weights are open, so you can run it yourself.
How can I use Qwen3 VL 235B A22B (Reasoning)?›
Through the Alibaba (Qwen) API.
What is the context window of Qwen3 VL 235B A22B (Reasoning)?›
262K tokens.
When was Qwen3 VL 235B A22B (Reasoning) released? Is there a newer version?›
Released 2025-09-23. It has been superseded by Qwen3.5 397B A17B.
Is Qwen3 VL 235B A22B (Reasoning) open source?›
Yes, the weights are open.
Can I run Qwen3 VL 235B A22B (Reasoning) locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
What is Qwen3 VL 235B A22B (Reasoning)?›
Qwen3 VL 235B A22B (Reasoning) is an open-weights language model from Alibaba (Qwen), released Sep 2025.
What are the best alternatives to Qwen3 VL 235B A22B (Reasoning)?›
Ling 3.1 Flash 1019 (cheaper), GPT-6 Luna (better), MiMo v2.6 Pro (open weights).