text-gen 235B params
Qwen/Qwen3-235B-A22B-Instruct-2507-FP8
Qwen/Qwen3-235B-A22B-Instruct-2507-FP8
355,830 downloads
147 likes
Hardware Recommendations
Three tiers based on VRAM requirements: minimum for basic inference, comfortable for longer contexts, headroom for future model sizes.
Minimum $7,999
Apple Mac Studio (M4 Max, 128GB)
128GB unified 85W TDP
Runs 235B at Q4_K_M quantization with 4K context.
Comfortable $7,999
Apple Mac Studio (M4 Max, 128GB)
128GB unified 85W TDP
Runs 235B at Q5_K_M quantization with 8K+ context.
VRAM Requirements
| Quantization | VRAM | Use case |
|---|---|---|
| Q4_K_M | 143 GB | Basic inference, 4K context |
| Q5_K_M | 179 GB | Good quality, 8K+ context |
| Q8 | 284 GB | Near-lossless, 16K+ context |
| FP16 | 566 GB | Full precision, max quality |
Formula: params × bytes_per_param × 1.2 overhead + 2 GB base. Actual requirements vary by framework and context length.