Qwen3.6-35B-A3B
Released
2026-04-16
Last refreshed
2026-10-13
Status
Researched 137d ago
Open sourceCommercial use: permittedMultimodalCodingRAGAgentsLong contextVisionClassificationJSON / Tool use
Alibaba releases · 52 in the last 12 months · this family litChangelog →
Specifications
- Family
- Qwen3.6
- Released
- 2026-04-16
- Context
- 256K
- Parameters
- 35B
- Architecture
- Mixture of Experts
- Specialization
- code
- Openness
- Open source
- License
- Apache 2.0OSI-approvedPermitted
- Weights
- Available
- Code
- Unknown
Created by
Pricing
Output / 1M
$1.00
Input / 1M
$0.150
Cheapest of 3 routes · OpenRouter
About
Qwen3.6-35B-A3B is an open-weight multimodal MoE model with 35B total parameters and 3B activated per token, released April 2026. It features a hybrid architecture combining Gated DeltaNet linear attention and standard Gated Attention with 256 total experts (8 routed + 1 shared), and includes a vision encoder for image and video understanding. Optimized for agentic coding, long-context reasoning, and visual tasks; supports 256K native context (extensible to ~1M via YaRN) with integrated thinking mode for multi-turn agent interactions.
Top use-case fit: coding, agents, and build tasks
Coding
Q/$ CAgents
Q/$ BProviders(3)
Compare all 3Compare API pricing across 3 tracked provider routes for input and output tokens, batch, and cached reads when available.
| Provider | Input / 1M | Output / 1M | Cache | Route |
|---|---|---|---|---|
| OpenRouter | $0.150 | $1.00 | - | Serverless |
| Novita AI | $0.248 | $1.49 | - | Serverless |
| Yandex AI Studio | $1.64 | $2.46 | read $0.4098 | Serverless |
Capabilities
VisionMultimodalJSON / Tool use
Benchmark peer barsfor Coding
Benchmark scores(8)
Scores are benchmark-specific and are direction-aware: the same numeric gap can mean very different outcomes across suites. Use the leaderboard context and this model's provider route to decide whether the winning margin is meaningful for your workload.
| Benchmark | Score | Version | Evaluation | Source |
|---|---|---|---|---|
| 73.4 | From official model card (resolved)Observed 2026-06-07 | — | ||
| 49.5 | —Observed 2026-04-16 | — | ||
| 80.4 | v6Observed 2026-04-16 | — | ||
| 85.2 | From official HuggingFace model card (accuracy)Observed 2026-06-07 | — | ||
| 86.0 | diamondObserved 2026-04-16 | — | ||
| 86.4 | miniObserved 2026-04-16 | — | ||
| 75.3 | LLM-Stats aggregatorObserved 2026-06-07 | — | ||
| 92.7 | AIME 2026 (accuracy)Observed 2026-06-07 | — |
Migration checks
No linked migration route is available for this model yet.
Rankings & picks(1)
Compare Qwen3.6-35B-A3B with other models
Comparison and alternatives
Browse all comparisons →Qwen3.6-35B-A3B vs Qwen3.6-27BQwen3.6-35B-A3B vs DeepSeek V4 FlashQwen3.6-35B-A3B vs DeepSeek V4 ProQwen3.6-35B-A3B vs Qwen3.5-122B-A10BQwen3.6-35B-A3B vs Qwen3-MaxQwen3.6-35B-A3B vs Claude Sonnet 4.6Qwen3.6-35B-A3B vs DeepSeek R1Qwen3.6-35B-A3B vs Qwen3.5-27BQwen3.6-35B-A3B vs Llama 3 70B InstructQwen3.6-35B-A3B vs Qwen3.6 Max PreviewQwen3.6-35B-A3B vs GLM-5.1Qwen3.6-35B-A3B vs Claude Opus 4.6Qwen3.6-35B-A3B vs Gemini 2.5 FlashQwen3.6-35B-A3B vs GPT-5.5Qwen3.6-35B-A3B vs Grok 4Qwen3.6-35B-A3B vs Llama 3.1 70B Instruct
Show all 36 popular comparisonssorted by 7-day search impressions
Qwen3.6-35B-A3B vs Qwen3.5-Plus159Qwen3.6-35B-A3B vs Llama 3 8B Instruct150Qwen3.6-35B-A3B vs Claude 3.7 Sonnet145Qwen3.6-35B-A3B vs DeepSeek V3129Qwen3.6-35B-A3B vs Claude Sonnet 4.5124Qwen3.6-35B-A3B vs DeepSeek V3.2124Qwen3.6-35B-A3B vs GPT-5.4120Qwen3.6-35B-A3B vs Claude Opus 4.5112Qwen3.6-35B-A3B vs Claude Opus 4.795Qwen3.6-35B-A3B vs Grok-390Qwen3.6-35B-A3B vs Mixtral 8x7B89Qwen3.6-35B-A3B vs GPT-5.280Qwen3.6-35B-A3B vs Kimi K2.579Qwen3.6-35B-A3B vs Qwen2.5-72B-Instruct63Qwen3.6-35B-A3B vs DeepSeek R1 052858Qwen3.6-35B-A3B vs DeepSeek V3.151Qwen3.6-35B-A3B vs Trinity-Large-Thinking50Qwen3.6-35B-A3B vs Qwen3.5-397B-A17B45Qwen3.6-35B-A3B vs Phi-4 14B34Qwen3.6-35B-A3B vs GPT-5.4 Pro34Qwen3.6-35B-A3B vs GLM-532Qwen3.6-35B-A3B vs Qwen3.5-35B-A3B32Qwen3.6-35B-A3B vs Gemini 2.5 Pro32Qwen3.6-35B-A3B vs Mistral Large 2 (2407)32Qwen3.6-35B-A3B vs Llama 3.2 1B Instruct31Qwen3.6-35B-A3B vs Phi-3 Mini 4k21Qwen3.6-35B-A3B vs o318Qwen3.6-35B-A3B vs Qwen3.5-Flash18Qwen3.6-35B-A3B vs Phi-4 Mini Flash Reasoning17Qwen3.6-35B-A3B vs Trinity-Large-Preview15Qwen3.6-35B-A3B vs DeepSeek R1 Lite14Qwen3.6-35B-A3B vs Llama 2 13B Chat11Qwen3.6-35B-A3B vs Gemini 2.5 Pro Preview 05-068Qwen3.6-35B-A3B vs Gemma 7B Instruct7Qwen3.6-35B-A3B vs Phi 4 Multimodal Instruct3Qwen3.6-35B-A3B vs o3 Mini1
Created by
Pricing
Output / 1M
$1.00
Input / 1M
$0.150
Cheapest of 3 routes · OpenRouter