[whatmodel.sh](../../README.md) / Qwen3 30B-A3B (Apr 2025, hybrid)
# Qwen3 30B-A3B (Apr 2025, hybrid)
Alibaba · released 2025-04 · 30.5B params (3.3B active) · apache-2.0
**score 55.3** ±0.4 · #274 of 732 · from 112 benchmark results across 16 sources
Score within the [LMArena (text)](https://whatmodel.sh/?set=lmarena-text&score=set) comparison set only.
## What it's good at
● marks Qwen3 30B-A3B (Apr 2025, hybrid) among every model on that task or test, worst to best.
task vs the others rank counts tests
agent tasks ▁▁▃▆●▄▅▅▄▄▂▁ #218 of 313 22% 6
coding ▁▁▁▃▅▇●▇▅▄▃▁ #254 of 469 10% 3
chat & writing ▁▁▁▃▆▄▆●▆▇▇▃ #198 of 402 11% 5
math ▁▁▃▅█●▄█▄▂▁▁ #197 of 431 15% 4
agent tasks · #218 of 313 · 22% of its score
Beats [o1 (2024-12-17)](o1-2024-12-17.md), [o3-mini](o3-mini.md), [DeepSeek R1 (Jan '25)](deepseek-r1.md) here, though they rank higher overall. Trails [SmolLM3 3B](smollm3-3b.md), [Jamba 1.5 Mini](jamba-1-5-mini.md), [Mellum2 12B-A2.5B Instruct](mellum2-12b-a2-5b-instruct.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
τ²-bench Telecom (maker) 22.8% ●·█········█ beats 0/6 very hard · avg 9.8% 9.4% sharp · several runs
τ-bench Retail (maker) 38.3% ▁····▃·▂●█▂▆ beats 12/28 medium · avg 36% 3.8% at its level · several runs
τ²-bench Retail (maker) 31.6% ▁····●▁▃▆█▃▁ beats 1/44 hard · avg 28% 2.9% at its level · small run · several runs
τ-bench Airline (maker) 18.0% ▁▁▁·●▄▁█▇▁▁▇ beats 4/28 hard · avg 28% 2.5% at its level · small run · several runs
τ²-bench Airline (maker) 18.0% ▁▁·●▁▃▂▁█▅▂▃ beats 2/42 hard · avg 17% 2.2% small run · several runs
BFCL v3 65.7% ▁▁▁▁▁▃▅█▅●▁▁ beats 99/114 medium · avg 51% 1.6% blunt · at its level · several runs
coding · #254 of 469 · 10% of its score
Beats [Qwen3-VL 32B Instruct](qwen3-vl-32b-instruct.md), [Solar Open 100B (102B-A12B)](solar-open-100b.md), [GLM-4.7-Flash](glm-4-7-flash.md) here, though they rank higher overall. Trails [QwQ-32B-Preview](qwq-32b-preview.md), [InternLM2.5 20B Chat](internlm2-5-20b-chat.md), [Ministral 3 3B Reasoning 2512](ministral-3-3b-reasoning.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
Aider polyglot (225 exercises, tw… 24.4% ▂▁▁▃▆●▄▆█▃▄▂ beats 24/71 very hard · avg 8.3% 5.2% sharp · small run · several runs
LiveCodeBench v6 (Feb–May 2025) (… 57.4% ▁▁▃▁█●▄▇▅▅▃█ beats 12/38 medium · avg 43% 3.1% at its level · small run · several runs
MultiPL-E (maker) 74.6% ▃··▄·▅▅▇▄●▃▄ beats 21/31 easy · avg 66% 1.3% at its level · small run · several runs
chat & writing · #198 of 402 · 11% of its score
Beats [Gemma 2 9B IT SimPO](gemma-2-9b-simpo.md), [Granite 4.2 8B](granite-4-2-8b.md), [Mistral Small 4 (119B-A6.5B)](mistral-small-4.md) here, though they rank higher overall. Trails [LFM2.5 1.2B Instruct](lfm2-5-1-2b-instruct.md), [LFM2.5 1.2B Thinking](lfm2-5-1-2b-thinking.md), [LFM2.5 8B-A1B](lfm2-5-8b-a1b.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
Arena-Hard v2 aggregate (maker) 36.3% ▁▁··●▃▁▃█▁▁▇ beats 4/22 hard · avg 27% 4.6% sharp · at its level · small run · seve…
LMArena (text) 1326.7 ▁▁▁▂▃▃▃●▆▇█▃ beats 164/385 avg 1260 2.1% blunt
IFEval (aggregation unspecified) … 86.5% ▁▂▄▂▂▄▆●█▅▂▁ beats 83/145 easy · avg 76% 1.9% small run · several runs
Arena-Hard v2.0 hard prompts (sty… 33.9% ▂·▃▅▅·●█▆▂▂▂ beats 10/22 very hard · avg 12% 1.9% sharp
Arena-Hard v2.0 creative writing … 34.9% ▆▂▂▆●▂▃█▂▃·▂ beats 10/22 hard · avg 19% 0.5% sharp · at its level · noisy
math · #197 of 431 · 15% of its score
Beats [MiniMax-M2.1](minimax-m2-1.md), [DeepSeek V3 (Dec '24)](deepseek-v3.md), [Qwen3 235B-A22B (Apr 2025, hybrid)](qwen3-235b-a22b.md) here, though they rank higher overall. Trails [Mixtral 8x7B v0.1 (base)](mixtral-8x7b-v0-1.md), [Phi-4-mini-reasoning 3.8B](phi-4-mini-reasoning.md), [DeepSeek R1 Distill Qwen 7B](deepseek-r1-distill-qwen-7b.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
AIME 2025 (maker) 70.9% ▁·▁▁▁▂▅▆●▇▅▁ beats 66/133 medium · avg 55% 5.7% sharp · at its level · small run · seve…
HMMT 2025 (edition unspecified) (… 49.8% ▃●▇▅▇▅▁·▁··▁ beats 8/32 hard · avg 34% 5.7% sharp · at its level · small run · seve…
PolyMATH (maker) 46.1% ▁···▃▁·●█▄▄▄ beats 7/23 hard · avg 35% 2.4% at its level · small run · several runs
OTIS mock AIME 2024–2025 62.8% ▃▃▄▃▃▃●▇█▃▂▅ beats 91/197 very hard · avg 7.2% 1.1% sharp · noisy
other tests (22)
test result vs the others in its run difficulty counts why
GPQA (subset unspecified) (maker) 65.8% ▂▁▂▂▄▃▆▃▅●▃▃ beats 70/106 hard · avg 51% 6.7% sharp · at its level · small run · seve…
MMLU-Pro (maker) 78.5% ▁▁·▁▁▁▃▂▄●█▁ beats 119/188 medium · avg 58% 4.1% at its level · small run · several runs
MMLU-Redux (maker) 89.5% ▁▁▃▁▂▃▄▅●▅█▄ beats 34/63 easy · avg 78% 3.6% small run · several runs
INCLUDE (maker) 71.9% ▂····▂▂▄▄●█▄ beats 19/36 medium · avg 58% 3.4% at its level · small run · several runs
GPQA Diamond 61.7% ▁▁▁▁▂▄▆●▅▇█▅ beats 89/211 hard · avg 39% 3.1% sharp · at its level
SuperGPQA (maker) 51.8% ▂▁▂▂▅▆▃●▇▇█▃ beats 21/46 hard · avg 34% 2.3% at its level · small run · several runs
MMLU-ProX (maker) 73.1% ▁·▁▁▁▂▁●▃█▅▄ beats 11/36 medium · avg 57% 2.2% at its level · small run · several runs
Creative Writing v3 (maker) 79.1% ▂····▂●··▆█▆ beats 4/14 easy · avg 71% 2.2% small run · several runs
WritingBench (maker) 77.0% ▁·●▂▁··▅█▂▂▁ beats 1/20 easy · avg 70% 2.0% small run · several runs
LiveBench 2024-11-25 (maker) 74.3% ▄··▂·▂·●█▆▂▂ beats 4/13 medium · avg 63% 1.9% at its level · small run · several runs
Multi-IF (aggregation unspecified… 72.2% ▁▁▁▁▁▁▃●█▄▁▂ beats 15/41 medium · avg 63% 1.7% at its level · small run · several runs
OJBench (maker) 20.7% ▁···▁▃·▅●▅██ beats 9/26 very hard · avg 13% 1.4% small run · several runs
CPTUB average (SpeakLeash, 0-5 ju… 70.7% ▁▂▂▂▃█▃●▅▄▂▃ beats 55/95 medium · avg 65% 1.3% at its level
EQ-Bench 3 (emotional intelligenc… 1082.9 ▁·▂▁●▅▅▃█▄▁▁ beats 10/62 avg 1034 1.3% typical
BRIDGE clinical text, zero-shot a… 40.9% ▁▁▁▁▁▂▂▃▄▆●▃ beats 80/97 hard · avg 31% 1.2% blunt · at its level
KINA (Knowledge Index of Noah's A… 24.3% ▁▁▂▁▃●█▆▄▅▂▃ beats 8/39 very hard · avg 21% 1.0% typical
Open Persian LLM Leaderboard v2 a… 55.2% ▂▄▂▃█●▃▆▃▅▂▂ beats 42/81 medium · avg 57% 0.9% blunt · at its level
ZebraLogic (maker) 33.2% ●·▁▁▁▇·▄▄▁▄▁ beats 2/24 medium · avg 45% 0.6% at its level · noisy · small run · seve…
Open PT-LLM Leaderboard average (… 76.8% ▃▁▁▁▁▁▁▁▂█●▁ beats 237/276 medium · avg 72% 0.4% at its level · noisy
EuroEval HellaSwag (English), tes… 81.7% ▃·▁▁▂▁▂▅█●▄▁ beats 76/100 medium · avg 67% 0.3% sharp · at its level · noisy
EuroEval Life in the UK (English … 94.8% ▁▁▁▁▁▁▁▂▄█▄● beats 140/146 easy · avg 82% 0.2% noisy
Chess Puzzles (Epoch) 4.0% ●▁▁▁▁▂▃▃▃▂▁▁ beats 3/137 very hard · avg 5.8% 0.2% sharp · too hard for it · noisy
source results share of the score
Qwen/Qwen3-4B-Instruct-2507 model card … 25 21% ▇▇▇
Qwen/Qwen3-30B-A3B-Thinking-2507 model … 23 21% ▇▇
Qwen/Qwen3-4B-Thinking-2507 model card … 23 20% ▇▇
Qwen/Qwen3-30B-A3B-Instruct-2507 model … 25 19% ▇▇
Epoch AI 3 4% ▇
Qwen (submitted / maker-reported) 1 2% ▇
Arena-Hard-Auto v2.0-Preview leaderboard 2 2% ▇
LMArena leaderboard dataset 1 2% ▇
SpeakLeash CPTUB 1 1% ▇
EQ-Bench 3 canonical leaderboard results 1 1% ▇
BRIDGE Medical Leaderboard 1 1% ▇
Knowledge Index of Noah's Ark 1 1% ▇
Open Persian LLM Leaderboard v2 1 1% ▇
EuroEval leaderboards 2 1% ▇
Berkeley Function Calling Leaderboard r… 1 0% ▇
Open Portuguese LLM Leaderboard 1 0% ▇
prior: pull toward the middle 0% ▇
weighted average = 55.3
Licences: Qwen/Qwen3-4B-Instruct-2507 model card ([apache-2.0](../../licenses/Qwen--Qwen3-4B-Instruct-2507.txt)) · Qwen/Qwen3-30B-A3B-Thinking-2507 model card ([apache-2.0](../../licenses/Qwen--Qwen3-30B-A3B-Thinking-2507.txt)) · Qwen/Qwen3-4B-Thinking-2507 model card ([apache-2.0](../../licenses/Qwen--Qwen3-4B-Thinking-2507.txt)) · Qwen/Qwen3-30B-A3B-Instruct-2507 model card ([apache-2.0](../../licenses/Qwen--Qwen3-30B-A3B-Instruct-2507.txt)) · Epoch AI ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)) · Arena-Hard-Auto v2.0-Preview leaderboard ([Apache-2.0](../../licenses/arena-hard-auto-Apache-2.0.txt)) · LMArena leaderboard dataset ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)) · SpeakLeash CPTUB ([MIT](../../licenses/hf-leaderboard-spaces-MIT.txt)) · EQ-Bench 3 canonical leaderboard results ([MIT](../../licenses/eqbench3-MIT.txt)) · BRIDGE Medical Leaderboard ([MIT](../../licenses/bridge-medical-leaderboard-MIT.txt)) · Knowledge Index of Noah's Ark ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)) · Open Persian LLM Leaderboard v2 ([Apache-2.0](../../licenses/hf-leaderboard-spaces-Apache-2.0.txt)) · EuroEval leaderboards ([MIT](https://github.com/EuroEval/leaderboards/blob/1a77f1d2f8ba1ba05ab73287fb090d76bf41b2e0/LICENSE)) · Berkeley Function Calling Leaderboard result archive ([Apache-2.0](../../licenses/bfcl-result-archive-v4-Apache-2.0.txt)) · Open Portuguese LLM Leaderboard ([Apache-2.0](../../licenses/hf-leaderboard-spaces-Apache-2.0.txt))
## How to run it
### Through an API
api (Alibaba Cloud Model Studio) · $0.35/M blended · $0.2 in / $0.8 out per 1M tokens · [alibabacloud.com](https://www.alibabacloud.com/help/en/model-studio/text-generation-model)
[price source](https://www.alibabacloud.com/help/en/model-studio/model-pricing): USD per 1M tokens, Singapore (International), qwen3-30b-a3b (hybrid; non-thinking output price; thinking-mode output is $2.4), single tier: input $0.2, output $0.8. Batch and context-cache discounts, higher tiers and free quota excluded. Observed 2026-09-28. · [context source](https://huggingface.co/Qwen/Qwen3-30B-A3B/blob/ad44e777bcd18fa416d9da3bd8f70d33ebb85d39/README.md): Native/default model-card context. Any YaRN extension must be configured separately; hosted provider limits may differ.
### llama.cpp (recommended)
1. Install llama.cpp. Homebrew and winget installs name the commands llama-cli and llama-server instead. [Other ways to install](https://github.com/ggml-org/llama.cpp/blob/master/docs/install.md)
```sh
# macOS / Linux
curl -LsSf https://llama.app/install.sh | sh
# Windows
irm https://llama.app/install.ps1 | iex
```
2. Run Qwen3 30B-A3B (Apr 2025, hybrid) (Q4_K_M, needs ~18.6 GB, from [huggingface.co/Qwen](https://huggingface.co/Qwen/Qwen3-30B-A3B-GGUF/tree/e4d4bafdfb96a411a163846265362aceb0b9c63a))
```sh
llama cli -hf Qwen/Qwen3-30B-A3B-GGUF:Q4_K_M
```
3. Or start a local server and open the address it prints (http://localhost:9931)
```sh
llama serve -hf Qwen/Qwen3-30B-A3B-GGUF:Q4_K_M
```
Other versions: Q8_0 (~32.5 GB) `llama cli -hf Qwen/Qwen3-30B-A3B-GGUF:Q8_0`
### LM Studio
1. Install LM Studio. Open the app once before using the lms command. [Download it (macOS / Windows / Linux)](https://lmstudio.ai/download)
2. Run Qwen3 30B-A3B (Apr 2025, hybrid) (Q4_K_M, needs ~18.6 GB, from [huggingface.co/Qwen](https://huggingface.co/Qwen/Qwen3-30B-A3B-GGUF/tree/e4d4bafdfb96a411a163846265362aceb0b9c63a))
```sh
lms get https://huggingface.co/Qwen/Qwen3-30B-A3B-GGUF@Q4_K_M -y
```
3. Then run lms ls to see its name, and lms chat followed by that name
Other versions: Q8_0 (~32.5 GB) `lms get https://huggingface.co/Qwen/Qwen3-30B-A3B-GGUF@Q8_0 -y` · 4bit (~16.2 GB, Mac only) `lms get https://huggingface.co/Qwen/Qwen3-30B-A3B-MLX-4bit -y` · 8bit (~31.5 GB, Mac only) `lms get https://huggingface.co/Qwen/Qwen3-30B-A3B-MLX-8bit -y`
### Ollama
1. Install Ollama. [Download it (macOS / Windows)](https://ollama.com/download)
```sh
# Linux
curl -fsSL https://ollama.com/install.sh | sh
```
2. Run Qwen3 30B-A3B (Apr 2025, hybrid) (Q4_K_M, needs ~18.8 GB, from [ollama.com](https://ollama.com/library/qwen3/tags))
```sh
ollama run qwen3:30b-a3b-q4_K_M
```
Other versions: Q8_0 (~37.1 GB) `ollama run qwen3:30b-a3b-q8_0`
### MLX (Mac)
1. Install MLX (Mac). Apple silicon Macs only, text only.
```sh
# with uv
uv tool install mlx-lm
# or
pip install mlx-lm
```
2. Run Qwen3 30B-A3B (Apr 2025, hybrid) (4bit, needs ~16.2 GB, Mac only, from [huggingface.co/Qwen](https://huggingface.co/Qwen/Qwen3-30B-A3B-MLX-4bit/tree/4e2776a4cc73a8a251d0b797010a5b07fd541a3e))
```sh
mlx_lm.chat --model Qwen/Qwen3-30B-A3B-MLX-4bit
```
3. Or start an OpenAI-compatible server at http://localhost:8080/v1
```sh
mlx_lm.server --model Qwen/Qwen3-30B-A3B-MLX-4bit
```
Other versions: 8bit (~31.5 GB, Mac only) `mlx_lm.chat --model Qwen/Qwen3-30B-A3B-MLX-8bit`
Model repository and documentation: [huggingface.co/Qwen](https://huggingface.co/Qwen/Qwen3-30B-A3B)
[interactive version](https://whatmodel.sh/model/qwen3-30b-a3b) · [JSON](https://whatmodel.sh/api?format=json&all=1)