Model Detail human-review-recommended

Qwen3-Coder 30B-A3B

Alibaba's Qwen3-Coder 30B with MoE architecture. Top coding model in <30B class.

AlibabaQWEN3apache-2.02025-07-23
Parameters
30.5B
3.3B active
Context window
262K
Long context
Architecture
decoder only transformer
moe
Quality score
89
Planner signal

Task Fit

Code AgentSupported

Tool use, repo work, terminal workflows, and coding benchmarks.

CodeSupported

Code generation, debugging, refactoring, and benchmark signal.

ChatNot a fit

Not marked for chat in the current library.

RAGNot a fit

Not marked for rag in the current library.

VisionNot a fit

Not marked for vision in the current library.

Image GenerationNot a fit

Not marked for image generation in the current library.

Video GenerationNot a fit

Not marked for video generation in the current library.

VoiceNot a fit

Not marked for voice in the current library.

Source Confidence

Overallhigh · 94/100
ParametersReviewed / seeded
Task fitReviewed / seeded
MemorySeeded artifact
LicenseSource / seed
BenchmarksAvailable
Hardware fitCalculated

Variants and Quant Artifacts

Choose the artifact first; hardware fit follows from RAM, VRAM, format, and runtime.

4 artifacts
QuantFormatQualityMin RAMReco RAMRuntimeAction
Q4_K_Mggufbalanced24GB48GBollama, llama.cpp, lm-studio Plan with this
Q5_K_Mggufbalanced24GB48GBollama, llama.cpp, lm-studio Plan with this
Q8_0ggufhigh35GB48GBollama, llama.cpp, lm-studio Plan with this
FP16ggufhigh68GB48GBollama, llama.cpp, lm-studio Plan with this

Recommended Hardware

Cheapest That Works
NVIDIA RTX 3060 12GB
Compatible

Lowest estimated 5-year cost that can run this model.

12GB VRAM$1,570 / 5y
Best Value
Ryzen AI Max+ 395 128GB Mini PC
Compatible

Plenty of fast-memory headroom for this model.

128GB RAM$2,311 / 5y
Best Performance
NVIDIA RTX 6000 Blackwell 96GB
Compatible

Highest local performance signal among compatible hardware.

96GB VRAM$12,376 / 5y

Benchmarks

Humaneval89.2%
MMLU81%
Livecodebench74%

Source and Review

Hugging FaceQwen/Qwen3-Coder-30B-A3B-Instruct
Ollamaqwen3-coder:30b
Verificationhuman-review-recommended
Artifact sourceseeded
Default variantQwen3-Coder 30B-A3B Coder
Tool callingSupported

Similar Models