MPT-30B-Instruct needs ~59.0 GB VRAM. Gaudi 3 128GB has 128.0 GB. With Q5_K_M quantization, expect ~122 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
122.3 tok/s
TTFT
1583 ms
Safe context
8K
Memory
59.0 GB / 128.0 GB
The raw memory story may look fine, but the software ecosystem is still a constraint here.
Runtime ecosystem is narrower than CUDA
Intel GPUs can look attractive on memory per dollar, but local AI tooling, kernels, and model coverage are still broader and easier on CUDA today.
Prefer CUDA if you want the path of least resistance
If your goal is maximum runtime coverage, easier troubleshooting, and better support for new local AI releases, CUDA is usually still the safer upgrade path.
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Chat | A | Runs well | 122.3 tok/s | 863 ms | 8K |
| Coding | A | Runs well | 122.3 tok/s | 1583 ms | 8K |
| Agentic Coding | A | Runs well | 122.3 tok/s | 2302 ms | 8K |
| Reasoning | A | Runs well | 122.3 tok/s | 1871 ms | 8K |
| RAG | A | Runs well | 122.3 tok/s | 2878 ms | 8K |
How MPT-30B-Instruct (30B params) fits at each quantization level on Gaudi 3 128GB (128.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 11.7 GB | Low | B59 |
Q3_K_S | 3 | 14.7 GB | Low | B59 |
NVFP4 | 4 | 16.8 GB | Medium | B59 |
Q4_K_M | 4 | 18.3 GB | Medium | B60 |
Q5_K_M | 5 | 21.6 GB | High | B60 |
Q6_K | 6 | 24.6 GB | High | B60 |
Q8_0 | 8 | 32.1 GB | Very High | B61 |
F16Best for your GPU | 16 | 61.5 GB | Maximum | B66 |
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 99Your hardware
| Model | Params | Grade | Decode | Capabilities |
|---|---|---|---|---|
 Devstral 2 123B Instruct | 123B | S | 37.5 tok/s | |
 Qwen3-Coder 30B A3B Instruct | 30.5B | S | 391.6 tok/s | |
| Qwen 3.5 122B A10B | 122B | S | 104.1 tok/s | |
| Qwen 3.6 35B A3B | 35B | S | 329.1 tok/s | |
| Qwen 3.5 35B A3B | 35B | S | 357.9 tok/s |
Yes, Gaudi 3 128GB can run MPT-30B-Instruct with a A grade (Runs well). Expected decode speed: 122.3 tok/s.
MPT-30B-Instruct (30B parameters) requires approximately 59.0 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 Gaudi 3 128GB, MPT-30B-Instruct achieves approximately 122.3 tokens per second decode speed with a time-to-first-token of 1583ms using Q5_K_M quantization.
For coding workloads, MPT-30B-Instruct on Gaudi 3 128GB receives a A grade with 122.3 tok/s and 8K context.
On Gaudi 3 128GB, 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.
Prefer CUDA if you want the path of least resistance. If your goal is maximum runtime coverage, easier troubleshooting, and better support for new local AI releases, CUDA is usually still the safer upgrade path.
Often yes, if your goal is the easiest setup and the widest runtime support. Intel can offer attractive memory capacity, but CUDA still tends to win on tooling maturity, guides, kernels, and model coverage for local AI.
Paste this snippet into any page to show a live fit card.
<iframe src="https://willitrunai.com/embed/mpt-30b-instruct-on-gaudi-3-128gb" width="400" height="180" frameborder="0" style="border:none;border-radius:12px;overflow:hidden;" title="Will It Run AI — fit result"></iframe>
Preview: