Model benchmarks
ministral-3-3b-instruct-2512 local LLM performance
As of October 2026, ministral-3-3b-instruct-2512 runs at up to 93.7 tok/s for local inference (best of 4 community benchmark runs across 2 GPUs).
Model size
3B
Peak speed
93.7 tok/s
Average speed
63.7 tok/s
Avg PP
2828.9 tok/s
Min memory
5.2 GB
Max context
32,768 tokens
Avg output / run
7,075 tokens
Avg runtime / run
3m 45s
Avg quality
45.2
Benchmark runs
4
GPUs tested
2
Quality by task
Average LLM-judged quality (0–100) with the run-to-run spread shown as a P5–P95 band, overall and for each benchmark task, across all 4 runs. The low and high columns show how much the judge’s score varies between runs, and need at least two runs to display.
| Task | P5 (low) | Avg | P95 (high) |
|---|---|---|---|
| Overall | 40.5 | 45.2 | 50.1 |
| Agent Workflow | 29.9 | 38.4 | 45.6 |
| Code Generation | 26.6 | 30.7 | 36.0 |
| Role Play & Narrative | 60.8 | 65.3 | 69.2 |
| Research & Analysis | 36.5 | 46.4 | 62.0 |
Performance by hardware and tool
Every hardware/tool/quantization combination ministral-3-3b-instruct-2512 has been benchmarked on, ranked by peak token generation speed. Last updated October 2026.
| Hardware | Tool | Quant | Best tok/s | Avg tok/s | Memory | Context | Quality | Runs |
|---|---|---|---|---|---|---|---|---|
| NVIDIA GeForce RTX 3060 Laptop GPU | Bionic | — | 93.7 tok/s | 80.8 tok/s | 5.2 GB | 32,768 tokens | 47.0 | 3 |
| AMD Radeon Vega Series / Radeon Vega Mobile Series | Bionic | — | 12.6 tok/s | 12.6 tok/s | 13.5 GB | 8,192 tokens | 39.8 | 1 |
Benchmark runs
All 4 ministral-3-3b-instruct-2512 runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.
Frequently asked questions
- Is ministral-3-3b-instruct-2512 good for coding?
- In our benchmarks, ministral-3-3b-instruct-2512 scores 30.7/100 for coding. It runs at about 63.7 tok/s, so if you want more speed, Nex-N2.5-mini-IQ3_XXS is faster (~91.4 tok/s) and still scores well for coding (81.6/100).
- Is ministral-3-3b-instruct-2512 good for agentic (tool-using) tasks?
- In our benchmarks, ministral-3-3b-instruct-2512 scores 38.4/100 for agentic workflows. It runs at about 63.7 tok/s, so if you want more speed, Nex-N2.5-mini-APEX-Mini is faster (~88.2 tok/s) and still scores well for agentic workflows (91.3/100).
- How fast is ministral-3-3b-instruct-2512 for local inference?
- Across 4 community benchmark runs, ministral-3-3b-instruct-2512 reaches up to 93.7 tok/s and averages 63.7 tok/s, with the fastest results on NVIDIA GeForce RTX 3060 Laptop GPU.
- How much memory does ministral-3-3b-instruct-2512 need?
- The leanest observed configuration used about 5.2 GB of memory.
- Which tools have been used to run ministral-3-3b-instruct-2512?
- Benchmarks were submitted using Bionic. Results are community-contributed and updated as new runs arrive.