Nano GPT logo
NanoGPT
Provider logo

Qwen3 30B A3B

qwen/qwen3-30b-a3b
BackTry Model
Provider logo

Qwen3 30B A3B

qwen/qwen3-30b-a3b
BackTry Model

Qwen 3 30b A3B is a 30b model with 3 billion active parameters per pass. Supports switching between thinking and non thinking: trigger thinking with /think and /no_think anywhere in a prompt or system message to toggle chain-of-thought reasoning.

Added Feb 27, 2025

Model weights

Context Window

41.0K

Max Output

32.8K

Input Price (Auto)

$0.10/1M

Output Price (Auto)

$0.30/1M

Cache Read (Auto)

$0.050/1M

Benchmarks

Performance metrics and benchmarks

Sourced from Artificial Analysis.

Intelligence Index

6.6

Better than 19% of models compared

Reasoning

GPQA Diamond

Graduate-level scientific reasoning

51.5%

Better than 27% of models compared

HLE

Humanity's Last Exam

4.6%

Better than 26% of models compared

IFBench

Instruction-following benchmark

31.9%

Better than 19% of models compared

T²-Bench Telecom

Conversational AI agents in dual-control scenarios

22.2%

Better than 24% of models compared

AA-LCR

Long context reasoning evaluation

0.0%

Better than 6% of models compared

Coding

SciCode

Python programming for scientific computing

26.4%

Better than 31% of models compared

Terminal-Bench Hard

Agentic coding and terminal use

6.8%

Better than 37% of models compared

LiveCodeBench

Contamination-free coding benchmark

32.2%

Better than 39% of models compared

Math

AIME 2025

American Invitational Mathematics Examination 2025

21.7%

Better than 23% of models compared

AIME

American Invitational Mathematics Examination

26.0%

Better than 53% of models compared

Math-500

Diverse mathematical problem solving benchmark

86.3%

Better than 55% of models compared

Knowledge

MMLU-Pro

Professional and academic subject knowledge

71.0%

Better than 38% of models compared

Last updated Aug 16, 2026

Artificial Analysis

Providers

Auto routing is available for this model. Explicit provider selection is not available.

Loading provider options…

Compare Qwen3 30B A3B with similar models from the same provider or model family.

Qwen 3.6 35B A3B Uncensored

qwen/qwen3.6-35b-a3b-uncensored

Qwen 3.6 35B A3B Uncensored is an NVFP4 open-weight mixture-of-experts model LoRA-tuned for fewer refusals across chat, coding, tool use, and multimodal tasks.

Qwen3.6 35B A3B Thinking

Qwen/Qwen3.6-35B-A3B:thinking

Qwen3.6 35B A3B is a native vision-language MoE model with hybrid attention. Compared to Qwen3.5 35B A3B, Alibaba reports stronger agentic coding, mathematical and code reasoning, and better spatial understanding (including object localization and detection).

Qwen3.6 35B A3B

Qwen/Qwen3.6-35B-A3B

Qwen3.6 35B A3B is a native vision-language MoE model with hybrid attention. Compared to Qwen3.5 35B A3B, Alibaba reports stronger agentic coding, mathematical and code reasoning, and better spatial understanding (including object localization and detection).

Qwen3 Next 80B A3B (Instruct)

Qwen/Qwen3-Next-80B-A3B-Instruct

Based on the new Qwen3‑Next architecture (hybrid attention, highly sparse MoE, training‑stability optimizations, and multi‑token prediction), the Qwen3‑Next‑80B‑A3B‑Instruct model delivers extreme efficiency with only 3B active parameters per pass. It performs comparably to Qwen3‑235B‑A22B‑Instruct‑2507 and shows clear advantages on ultra‑long context tasks (up to 256K tokens).

Qwen3 Next 80B A3B (Thinking)

qwen/qwen3-next-80b-a3b-thinking

Qwen3 Next 80B A3B (Thinking)

Qwen 3.8 27B Uncensored

qwen/qwen3.8-27b-uncensored

Qwen 3.8 27B Uncensored is an NVFP4 open-weight multimodal model LoRA-tuned for fewer refusals across chat, coding, tool use, and long-context work.