QwenQwen3.5-122B-A10B on NVIDIA Grace Blackwell Ultra GB300
Yes — Grace Blackwell Ultra GB300 runs Qwen3.5-122B-A10B excellently at Q8_0 — 200 tok/s. 288 GB VRAM with plenty of headroom.
Model Size
122B
Device VRAM
288 GB
Bandwidth
8000 GB/s
Quantization
Q8_0
Performance by Quantization
OwnRig currently has one published compatibility entry for Qwen3.5-122B-A10B 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 | 200 tok/s | 35ms | ✓ Yes | Excellent | estimated |
Notes
Q8_0
Full quality Q8_0 with 223GB headroom. MoE: only 10B params active per token — faster than total size suggests.
About Qwen3.5-122B-A10B
Qwen3.5-122B-A10B (122B) is a chat, coding, ai coding, reasoning, multi-purpose model. Mixture-of-Experts with 122B total parameters but only 10B active per token. Runs on far less VRAM than the total size suggests. Qwen's best open-weight MoE in the medium tier; excels at knowledge-heavy and vision tasks. Native multimodal. Apache 2.0 licensed.
View all Qwen3.5-122B-A10B 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 Qwen 3 32B and MoE benchmarks with size/architecture scaling. Reference hardware source: github.com (2026-03-24)
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.