Qwen3 Coder 30B A3B Instruct MLX 8bit — Hardware Requirements & GPU Compatibility
ChatCode
An MLX 8-bit quantized version of Alibaba's Qwen3 Coder 30B A3B Instruct, converted by LM Studio Community for Apple Silicon Macs. Compared to the 5-bit variant, this 8-bit quantization retains more of the original model's precision and output quality, at the cost of higher memory usage. This mixture-of-experts coding model with 30.5 billion total parameters is well suited for developers on Apple Silicon who want near-full-quality code generation and are willing to dedicate more memory to achieve it. Users with 48GB or more of unified memory will get the best experience, though it may fit on 32GB machines with some context length limitations.
182.3K downloads
13 likesSpecifications
- Publisher
- LM Studio Community
- Family
- Qwen
- Parameters
- 30.5B
- Architecture
- Qwen3MoeForCausalLM
- Context Length
- 262,144 tokens
- Vocabulary Size
- 151,936
- Release Date
- 2025-07-31
- License
- Apache 2.0
Get Started
How Much VRAM Does Qwen3 Coder 30B A3B Instruct MLX 8bit Need?
Select a quantization to see compatible GPUs below.
| Quantization | Bits | VRAM | + Context | File Size | Quality |
|---|---|---|---|---|---|
| IQ2_XXS | 2.20 | 8.8 GB | 21.6 GB | 8.40 GB | Importance-weighted 2-bit, extreme compression — significant quality loss |
| IQ2_M | 2.70 | 10.7 GB | 23.5 GB | 10.30 GB | Importance-weighted 2-bit, medium |
| IQ3_XXS | 3.10 | 12.2 GB | 25.0 GB | 11.83 GB | Importance-weighted 3-bit |
| Q2_K | 3.40 | 13.4 GB | 26.2 GB | 12.98 GB | 2-bit quantization with K-quant improvements |
| Q3_K_S | 3.50 | 13.8 GB | 26.5 GB | 13.36 GB | 3-bit small quantization |
| Q3_K_M | 3.90 | 15.3 GB | 28.1 GB | 14.88 GB | 3-bit medium quantization |
| Q4_0 | 4.00 | 15.7 GB | 28.4 GB | 15.27 GB | 4-bit legacy quantization |
| IQ4_XS | 4.30 | 16.8 GB | 29.6 GB | 16.41 GB | Importance-weighted 4-bit, compact |
| Q4_1 | 4.50 | 17.6 GB | 30.4 GB | 17.17 GB | 4-bit legacy quantization with offset |
| Q4_K_S | 4.50 | 17.6 GB | 30.4 GB | 17.17 GB | 4-bit small quantization |
| IQ4_NL | 4.50 | 17.6 GB | 30.4 GB | 17.17 GB | Importance-weighted 4-bit, non-linear |
| Q4_K_M | 4.80 | 18.7 GB | 31.5 GB | 18.32 GB | 4-bit medium quantization — most popular sweet spot |
| Q5_K_S | 5.50 | 21.4 GB | 34.2 GB | 20.99 GB | 5-bit small quantization |
| Q5_K_M | 5.70 | 22.1 GB | 34.9 GB | 21.75 GB | 5-bit medium quantization — good quality/size tradeoff |
| Q6_K | 6.60 | 25.6 GB | 38.4 GB | 25.19 GB | 6-bit quantization, very good quality |
| Q8_0 | 8.00 | 30.9 GB | 43.7 GB | 30.53 GB | 8-bit quantization, near-lossless |
Which GPUs Can Run Qwen3 Coder 30B A3B Instruct MLX 8bit?
Q4_K_M · 18.7 GBQwen3 Coder 30B A3B Instruct MLX 8bit (Q4_K_M) requires 18.7 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 25+ GB is recommended. Using the full 262K context window can add up to 12.8 GB, bringing total usage to 31.5 GB. 6 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti.
Runs great
— Plenty of headroom
Which Devices Can Run Qwen3 Coder 30B A3B Instruct MLX 8bit?
Q4_K_M · 18.7 GB21 devices with unified memory can run Qwen3 Coder 30B A3B Instruct MLX 8bit, including NVIDIA DGX H100, NVIDIA DGX A100 640GB, Mac Mini M4 Pro (24 GB).
Runs great
— Plenty of headroomRelated Models
Frequently Asked Questions
- How much VRAM does Qwen3 Coder 30B A3B Instruct MLX 8bit need?
Qwen3 Coder 30B A3B Instruct MLX 8bit requires 18.7 GB of VRAM at Q4_K_M, or 30.9 GB at Q8_0. Full 262K context adds up to 12.8 GB (31.5 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 30.5B × 4.8 bits ÷ 8 = 18.3 GB
KV Cache + Overhead ≈ 0.4 GB (at 2K context + ~0.3 GB framework)
KV Cache + Overhead ≈ 13.2 GB (at full 262K context)
VRAM usage by quantization
Q4_K_M18.7 GBQ4_K_M + full context31.5 GB- Can NVIDIA GeForce RTX 4090 run Qwen3 Coder 30B A3B Instruct MLX 8bit?
Yes, at Q5_K_M (22.1 GB) or lower. Higher quantizations like Q6_K (25.6 GB) exceed the NVIDIA GeForce RTX 4090's 24 GB.
- What's the best quantization for Qwen3 Coder 30B A3B Instruct MLX 8bit?
For Qwen3 Coder 30B A3B Instruct MLX 8bit, Q4_K_M (18.7 GB) offers the best balance of quality and VRAM usage. Q5_K_S (21.4 GB) provides better quality if you have the VRAM. The smallest option is IQ2_XXS at 8.8 GB.
VRAM requirement by quantization
IQ2_XXS8.8 GB~53%Q3_K_S13.8 GB~77%Q4_117.6 GB~88%Q4_K_M ★18.7 GB~89%Q5_K_S21.4 GB~92%Q8_030.9 GB~99%★ Recommended — best balance of quality and VRAM usage.
- Can I run Qwen3 Coder 30B A3B Instruct MLX 8bit on a Mac?
Qwen3 Coder 30B A3B Instruct MLX 8bit requires at least 8.8 GB at IQ2_XXS, which exceeds the unified memory of most consumer Macs. You would need a Mac Studio or Mac Pro with a high-memory configuration.
- Can I run Qwen3 Coder 30B A3B Instruct MLX 8bit locally?
Yes — Qwen3 Coder 30B A3B Instruct MLX 8bit can run locally on consumer hardware. At Q4_K_M quantization it needs 18.7 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.
- How fast is Qwen3 Coder 30B A3B Instruct MLX 8bit?
At Q4_K_M, Qwen3 Coder 30B A3B Instruct MLX 8bit can reach ~156 tok/s on AMD Instinct MI300X. On NVIDIA GeForce RTX 4090: ~35 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = (bandwidth GB/s ÷ model GB) × efficiency
Example: AMD Instinct MI300X → 5300 ÷ 18.7 × 0.55 = ~156 tok/s
Estimated speed at Q4_K_M (18.7 GB)
AMD Instinct MI300X~156 tok/sNVIDIA GeForce RTX 4090~35 tok/sNVIDIA H100 SXM~116 tok/sAMD Instinct MI250X~96 tok/sReal-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
- What's the download size of Qwen3 Coder 30B A3B Instruct MLX 8bit?
At Q4_K_M, the download is about 18.32 GB. The full-precision Q8_0 version is 30.53 GB. The smallest option (IQ2_XXS) is 8.40 GB.