DeepYard
Q

Qwen3 235B A22B

25

Qwen3-235B-A22B is a 235B parameter mixture-of-experts (MoE) model developed by ...

Open Sourcepaid

About

Qwen3-235B-A22B is a 235B parameter mixture-of-experts (MoE) model developed by Qwen, activating 22B parameters per forward pass. It supports seamless switching between a "thinking" mode for complex reasoning, math, and code tasks, and a "non-thinking" mode for general conversational efficiency. The model demonstrates strong reasoning ability, multilingual support (100+ languages and dialects), advanced instruction-following, and agent tool-calling capabilities. It natively handles a 32K token c

Details

Modalitytext
Context Window131.1K tokens
Release DateApr 28, 2025
API AvailableYes
Hostingself-hosted, api
Output Speed120 tokens/sec
Time to First Token600ms
Quality Index82/100
Coding Index75/100
Reasoning Index84/100

Benchmarks

Chatbot Arena ELO
1400/1500
MMLU-Pro
84.5/100
SWE-bench Verified
52/100
MATH-500
90.5/100
GPQA Diamond
72/100
HumanEval
92.5/100

Tags

open-weightapichat