Qwen 3.5 35B A3B
alibaba/qwen3.5-35b-a3b- Alibaba
- Family · Qwen3
- Released 24 Feb 2026
- Open weights
- Text · Image · Video → Text
- Tools · JSON · Vision
Not on OSMEXO yet
Spec & list price
Context256K
Max output80K
Input / 1M$0.14
Output / 1M$1.007.1× input
Scores · standing
Intelligence19.3#146/489
Coding37.0#98/196
Usage
Tokens / day1.8B▲ 49% 7d
Usage rank#166
Requests / day262.6K
Tool calls / req0.07
Per request · 30 days
Avg input3.8K0.3× median
Avg output2790.4× median
Cache hit11%0.2× median
Reasoning share49%0.9× median
Tool-call failures2.7%1.4× median
Consumption
Peak 10.1B 30 OctLow 860.2M 25 OctRank today #166
News & trends
184 mentions this week ▲ 36%Qwen3 family · videos, posts, articles · 30 days
Trending · 14 days · Qwen3 family
- Memdoor: A team of 5 Qwen3.8-27B models beats Fable 5 on LiveCodeBench Hard453 upvotes
- Memdoor: A team of 5 Qwen3.8-27B models beats Fable 5 on LiveCodeBench Hard416 points
- Neuron Explorer for Qwen3.5-4B: See every neuron at work, make targeted edits26 points
- New Card for Qwen3.8 27b and MiniMax H3632 upvotes
- Qwen 3.8 27B vs Strata Coder - Local LLM battle75.2K views
- DavidAU Qwen3.8 27B Turbo Brilliance 50X-Reasoning... tested - Local LLM setup87.7K views
News · latest from Qwen3
- Memdoor: A team of 5 Qwen3.8-27B models beats Fable 5 on LiveCodeBench Hardgithub.com · 31 Oct
- Neuron Explorer for Qwen3.5-4B: See every neuron at work, make targeted editsgithub.com · 26 Oct
- How to train your own Jev for $17Together AI · 23 Sep
Benchmarks
bar = share of the top score · Benchmark data: Artificial AnalysisSpeed · median across 7 providers, latest day
Output speed#146/33473 tok/s1.1× faster than median
Fastest provider127 tok/sbest of 7 providers
First token#1721.08 sabout median
#6 cheapest of 24 with a similar score
Value
Intelligence Index
Blended price per 1M tokens (log)
Family progress
Intelligence Index
- Family releases33.7
Technical details
- Tokenizer
- Qwen3
- License
- Apache 2.0
- Parameters
- 36B
- Moderation
- none
- Downloads · likes
- 1.7M · 1.6K
Request parameters 19
- tools
- tool_choice
- response_format
- structured_outputs
- reasoning
- include_reasoning
- temperature
- top_p
- top_k
- min_p
+9 more
- max_tokens
- stop
- seed
- frequency_penalty
- logit_bias
- logprobs
- presence_penalty
- repetition_penalty
- top_logprobs
When to use
Best for
- Cost-efficient agents with tool calling
- Instruction-following assistants
- Long document analysis
- Image and video understanding in apps
- Self-hosted multimodal serving (open weights)
Think twice for
- Image or video generation
- Hardest coding agent tasks
- Audio input or output
- Top-tier frontier reasoning
Research papers
- Learning to Look Beyond Screens: Transferable Visual Observation from GUI Grounding27 Oct
- Towards Architectural-Level Safety Control: Relocating Hazardous Knowledge into Removable Experts in Mixture-of-Experts Models19 Oct
- LEASH: Uncertainty-Aware Weak-to-Strong Escalation and Oversight for Computer-Use Agents15 Oct
- Do Your Own Research: Learning to Forecast by Learning to Search1 Oct
- Learn Now, Use Next, Trust Later: Prequential Test-Time Learning for LLM Agents28 Sep
- SkillGym: Internalizing Human Skills into LLMs for Real-World Problem Solving23 Sep
- Realize What Matters: Principled Context Representation for Large-Scale Reasoning23 Sep
- MemCalib: Benchmarking and Optimizing Memory Use in LLM Agents21 Sep
Same family
Qwen3- 7.2
- 6.0
- 6.9
- 6.1
- 5.2
- 13.1
Alternatives
other vendors · similar capability- 21.5
- 22.2
- S19.5
- 20.6
Want Qwen 3.5 35B A3B on OSMEXO?
Sign in to request it. Requests are reviewed; a request does not guarantee availability.