Ministral 3 8B needs ~12.1 GB VRAM. Tesla P40 24GB has 24.0 GB. With Q4_K_M quantization, expect ~42 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
45.0 tok/s
TTFT
4305 ms
Safe context
103K
Memory
12.1 GB / 24.0 GB
This setup is broadly balanced for this model.
Older PCIe generation
PCIe 3.0 is workable, but it compounds the penalty when you offload heavily or try to scale across multiple cards.
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Chat | A | Runs well | 41.8 tok/s | 2524 ms | 103K |
| Coding | A | Runs well | 41.8 tok/s | 4628 ms | 103K |
| Agentic Coding | A | Runs well | 41.8 tok/s | 6732 ms | 103K |
| Reasoning | A | Runs well | 41.8 tok/s | 5470 ms | 103K |
| RAG | A | Runs well | 41.8 tok/s | 8415 ms | 103K |
How Ministral 3 8B (8B params) fits at each quantization level on Tesla P40 24GB (24.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 3.1 GB | Low | A75 |
Q3_K_S | 3 | 3.9 GB | Low | A75 |
NVFP4 | 4 |
Copy-paste commands to run Ministral 3 8B on your machine.
Run
docker run --rm -it ghcr.io/ggerganov/llama.cpp:full \
--hf-repo "mistralai/Ministral-3-8B-Instruct-2512" \
--hf-file "Ministral-3-8B-Instruct-2512-Q4_K_M.gguf" \
-c 4096 -ngl 99Your hardware
| Model | Params | Grade | Decode | Capabilities |
|---|---|---|---|---|
| 27B | S | 13.4 tok/s | ||
Yes, Tesla P40 24GB can run Ministral 3 8B with a A grade (Runs well). Expected decode speed: 41.8 tok/s.
Ministral 3 8B (8B parameters) requires approximately 12.1 GB of memory with Q4_K_M quantization.
The recommended quantization for Ministral 3 8B is Q4_K_M, which balances quality and memory efficiency.
On Tesla P40 24GB, Ministral 3 8B achieves approximately 41.8 tokens per second decode speed with a time-to-first-token of 4628ms using Q4_K_M quantization.
For coding workloads, Ministral 3 8B on Tesla P40 24GB receives a A grade with 41.8 tok/s and 103K context.
On Tesla P40 24GB, Ministral 3 8B can safely use up to 103K tokens of context. The model's official context limit is 262K, 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/ministral-3-8b-on-tesla-p40-24gb" width="400" height="180" frameborder="0" style="border:none;border-radius:12px;overflow:hidden;" title="Will It Run AI — fit result"></iframe>
Preview:
4.5 GB |
| Medium |
| A76 |
Q4_K_M | 4 | 4.9 GB | Medium | A76 |
Q5_K_M | 5 | 5.8 GB | High | A76 |
Q6_K | 6 | 6.6 GB | High | A77 |
Q8_0 | 8 | 8.6 GB | Very High | A78 |
F16Best for your GPU | 16 | 16.4 GB | Maximum | A80 |
| 9B |
| S |
| 40 tok/s |
| 24B | S | 15 tok/s |
| 24B | S | 15 tok/s |
| 14B | S | 25.8 tok/s |