Qwen3.6 35B A3B
ModelOpen sourceActiveby Alibaba (Qwen) · family “qwen3b-a3b”
Qwen3.6-35B-A3B is an open-weight multimodal model from Alibaba Cloud with 35 billion total parameters and 3 billion active parameters per token. It uses a hybrid sparse mixture-of-experts architecture combining Gated... (description from the OpenRouter listing)
in: textin: imagein: videoout: textReasoningTool useStructured output
Epoch Capabilities Index
Not scored by Epoch AI
Released
27 Apr 2026
Context window
262K tokens
Max output
236K tokens
Input $ / 1M tokens
$0.15
Output $ / 1M tokens
$1.00
Cached input $ / 1M
$0.050
Each benchmark shown separately with its own source
Benchmark results (10)
| Benchmark | Domain | Score | vs best recorded | Setting | Run | Source |
|---|---|---|---|---|---|---|
| LMCA | agents | 29.7% | 44% | — | — | External ↗ |
| DTBench | reasoning | 73.9% | 75% | — | — | External ↗ |
| Mystery Game Puzzles | games | 22.0% ±4.2 | 26% | none | 27 Aug 2026 | Epoch ↗ |
| Surface Evolver Bench | science | 44.4% | 47% | — | — | External ↗ |
| FrontierMath-Tiers-1-3-v2-Private | math | 20.4% ±2.4 | 22% | none | 30 Aug 2026 | Eval log ↗ |
| Chess Puzzles | games | 26.0% ±4.4 | 36% | — | 7 Aug 2026 | Eval log ↗ |
| Terminal Bench | agents | 23.0% ±3.2 | 27% | agent: little-coder | 16 Apr 2026 | External ↗ |
| WeirdML | coding | 34.5% ±0.0 | 37% | — | — | External ↗ |
| OTIS Mock AIME 2024-2025 | math | 86.7% ±5.1 | 87% | — | 7 Aug 2026 | Eval log ↗ |
| GPQA diamond | science | 84.8% ±2.6 | 89% | none | 7 Aug 2026 | Eval log ↗ |
Source: Epoch AI Benchmarking Hub (CC BY 4.0). “External” rows are leaderboard results Epoch collects from third parties. Best reported setting per benchmark is shown.
API list price over time
Price history
$ per 1M tokens
6 recorded prices on 4 May 2026 (OpenRouter listing + Internet Archive snapshots); re-read on every data refresh, most recently 10h ago. Steps show when the price changed.
Events
- BenchmarkEpoch AI evaluates Qwen3.6 35B A3BEpoch AI Benchmarking Hub
- BenchmarkEpoch AI evaluates Qwen3.6 35B A3BEpoch AI Benchmarking Hub
- BenchmarkEpoch AI evaluates Qwen3.6 35B A3BEpoch AI Benchmarking Hub
- Open releaseMajorQwen releases Qwen3.6 35B A3BOpenRouter models API