HF Model & Benchmark Matrix

24,078 variants · 14,984 scores · 120 benchmarks
Home/Orgs/Qwen/Qwen/Qwen3-30B-A3B-Instruct-2507

Qwen/Qwen3-30B-A3B-Instruct-2507

family qwen · 11 variants · unknown

provider
Qwen
released
—
huggingface
https://huggingface.co/Qwen/Qwen3-30B-A3B-Instruct-2507
best scores
MMLU-Pro: 81.2MMLU: 90.4GPQA: 68.4AIME 2025: 46.6LiveCodeBench: 45.2IFEval: 82.3Chatbot Arena: 45.6BFCL (function calling): 64.7

Best scores per benchmark

benchmarkcategorybest score
MMLUknowledge90.4
IFEvalinstruction-following82.3
MMLU-Proknowledge81.2
GPQAreasoning68.4
BFCL (function calling)agentic64.7
AIME 2025math46.6
Chatbot Arenachat45.6
LiveCodeBenchcoding45.2

Variants (11)

variantkindquantbitsvRAMdlorgyear
Qwen3-30B-A3B-Instruct-2507originalbf1616.02,116,385Qwen2025
Qwen3-30B-A3B-Instruct-2507-FP8fp8fp88.0597,620Qwen2025
Qwen3-30B-A3B-Instruct-2507-GGUFggufgguf16.0140,989unsloth2025
Qwen3-30B-A3B-Instruct-2507-MLX-4bitmlxmlx24,478lmstudio-community2025
Qwen3-30B-A3B-Instruct-2507-MLX-8bitmlxmlx23,067lmstudio-community2025
Qwen3-30B-A3B-Instruct-2507-MLX-6bitmlxmlx22,848lmstudio-community2025
Qwen_Qwen3-30B-A3B-Instruct-2507-GGUFggufgguf2.010,089bartowski2025
Qwen3-30B-A3B-Instruct-2507-GGUFggufgguf3.09,800lmstudio-community2025
Qwen3-30B-A3B-Instruct-2507originalbf1616.04,197unsloth2025
Qwen3-30B-A3B-Instruct-2507-Q8_0-GGUFggufgguf8.0429ggml-org2025
Qwen3-30B-A3B-Instruct-2507-GGUFggufgguf16.0163Mungert2025
Hugging Face model & benchmark data scraped & matrix-ified. 24,078 variants · 14,984 scores · 120 benchmarks.