MPT-30B-Instruct needs ~65.0 GB VRAM. H100 NVL 188GB has 188.0 GB. With Q5_K_M quantization, expect ~298 tok/s.
Operating mode
Interactive favors responsiveness, while light API and scale-out lean harder on serving readiness. The fit stays the same, but the recommendation lens changes.
Current mode
Balanced
Balanced for general local use. Keeps the ranking neutral across personal and serving workflows.
Select quantization to explore
Fit status
Runs well
Decode
298.4 tok/s
TTFT
649 ms
Safe context
8K
Memory
65.0 GB / 188.0 GB
This setup is broadly balanced for this model.
No major red flags
This recommendation has enough memory headroom and acceptable estimated speed for the selected workload.
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Chat | B | Runs well | 298.4 tok/s | 354 ms | 8K |
| Coding | B | Runs well | 298.4 tok/s | 649 ms | 8K |
| Agentic Coding | A | Runs well | 298.4 tok/s | 944 ms | 8K |
| Reasoning | B | Runs well | 298.4 tok/s | 767 ms | 8K |
| RAG | A | Runs well | 298.4 tok/s | 1180 ms | 8K |
How MPT-30B-Instruct (30B params) fits at each quantization level on H100 NVL 188GB (188.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 11.7 GB | Low | B58 |
Q3_K_S | 3 | 14.7 GB | Low | B58 |
NVFP4 | 4 | 16.8 GB | Medium | B58 |
Q4_K_M | 4 | 18.3 GB | Medium | B58 |
Q5_K_M | 5 | 21.6 GB | High | B58 |
Q6_K | 6 | 24.6 GB | High | B59 |
Q8_0 | 8 | 32.1 GB | Very High | B60 |
F16Best for your GPU | 16 | 61.5 GB | Maximum | B63 |
Copy-paste commands to run MPT-30B-Instruct on your machine.
Run
docker run --rm -it ghcr.io/ggerganov/llama.cpp:full \
--hf-repo "mosaicml/mpt-30b-instruct" \
--hf-file "mpt-30b-instruct-Q5_K_M.gguf" \
-c 4096 -ngl 99Yes, H100 NVL 188GB can run MPT-30B-Instruct with a B grade (Runs well). Expected decode speed: 298.4 tok/s.
MPT-30B-Instruct (30B parameters) requires approximately 65.0 GB of memory with Q5_K_M quantization.
The recommended quantization for MPT-30B-Instruct is Q5_K_M, which balances quality and memory efficiency.
On H100 NVL 188GB, MPT-30B-Instruct achieves approximately 298.4 tokens per second decode speed with a time-to-first-token of 649ms using Q5_K_M quantization.
For coding workloads, MPT-30B-Instruct on H100 NVL 188GB receives a B grade with 298.4 tok/s and 8K context.
On H100 NVL 188GB, MPT-30B-Instruct can safely use up to 8K tokens of context. The model's official context limit is 8K, but available memory constrains the safe maximum.
Paste this snippet into any page to show a live fit card.
<iframe src="https://willitrunai.com/embed/mpt-30b-instruct-on-h100-nvl-188gb" width="400" height="180" frameborder="0" style="border:none;border-radius:12px;overflow:hidden;" title="Will It Run AI — fit result"></iframe>
Preview: