TPSTokens Per Second

Qwen3.6-35B-A3B-FP8

Alibaba / Qwen🇨🇳 China
active
Compare with other models →
Context window262K tokens

Version History

3.6minor

First open-weight release of Qwen3.6 series, featuring improved agentic coding capabilities, thinking preservation for context retention, and FP8 quantization. Built on community feedback to prioritize stability and real-world utility.

Benchmark Scores

Full leaderboard →
86.0%
GPQA
85.2%
MMLU-Pro
81.7%
MMMU
73.4%
SWE-bench Verified

Coverage

model release

Alibaba Qwen Releases 35B Parameter Qwen3.6-35B-A3B Model with 262K Native Context Window

Alibaba Qwen has released Qwen3.6-35B-A3B, a 35-billion parameter mixture-of-experts model with 3 billion activated parameters and a 262,144-token native context window extendable to 1,010,000 tokens. The model scores 73.4 on SWE-bench Verified and features FP8 quantization with performance metrics nearly identical to the original model.

2 min read
QwenAlibabaModel Release