DeepSeekDeepSeek Coder V2 Lite 16B on NVIDIA Grace Blackwell Ultra GB300
Yes — Grace Blackwell Ultra GB300 runs DeepSeek Coder V2 Lite 16B excellently at Q8_0 — 210 tok/s. 288 GB VRAM with plenty of headroom.
Model Size
15.7B
Device VRAM
288 GB
Bandwidth
8000 GB/s
Quantization
Q8_0
Performance by Quantization
OwnRig currently has one published compatibility entry for DeepSeek Coder V2 Lite 16B on NVIDIA Grace Blackwell Ultra GB300 at Q8_0. This is the best supported pairing we can stand behind today.
| Quantization | Speed | TTFT | Fits in VRAM | Rating | Confidence |
|---|---|---|---|---|---|
| Q8_0 | 210 tok/s | 55ms | ✓ Yes | Excellent | estimated |
Notes
Q8_0
Full quality. 271GB headroom.
About DeepSeek Coder V2 Lite 16B
DeepSeek Coder V2 Lite 16B (15.7B) is a coding, ai coding, ai building model. Mixture-of-Experts architecture: 15.7B total, ~2.4B active per token. Fast and capable code generation and completion. Inference speed belies the total param count. One of the best coding models for its effective size.
View all DeepSeek Coder V2 Lite 16B hardware options →About NVIDIA Grace Blackwell Ultra GB300
NVIDIA Grace Blackwell Ultra GB300 has 288 GB at 8000 GB/s. Street price: $30,000.
See all models NVIDIA Grace Blackwell Ultra GB300 can run →Estimate method: Estimated from GB300 bandwidth ratio vs RTX 4090. Reference hardware source: dell.com (2026-03-23)
Performance varies by driver version, inference engine, quantization method, context length, and system configuration. Figures shown are estimates based on community benchmarks and may not reflect your exact setup. Product names are trademarks of their respective owners. OwnRig is independent and not affiliated with any hardware or AI model provider.