Model benchmarks
ministral-1b-base local LLM performance
As of October 2026, ministral-1b-base runs at up to 233.5 tok/s for local inference (best of 5 community benchmark runs across 1 GPU).
Model size
1B
Peak speed
233.5 tok/s
Average speed
230.4 tok/s
Avg PP
17889.9 tok/s
Min memory
2.7 GB
Max context
65,536 tokens
Avg output / run
1,610 tokens
Avg runtime / run
9s
Avg quality
0.4
Benchmark runs
5
GPUs tested
1
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 5 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 | 0.0 | 0.4 | 1.0 |
| Agent Workflow | 0.0 | 1.8 | 4.0 |
| Code Generation | 0.0 | 0.0 | 0.0 |
| Role Play & Narrative | 0.0 | 0.0 | 0.0 |
| Research & Analysis | 0.0 | 0.0 | 0.0 |
Performance by hardware and tool
Every hardware/tool/quantization combination ministral-1b-base 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 | — | 233.5 tok/s | 230.4 tok/s | 2.7 GB | 65,536 tokens | 0.4 | 5 |
Benchmark runs
All 5 ministral-1b-base runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.
Frequently asked questions
- How fast is ministral-1b-base for local inference?
- Across 5 community benchmark runs, ministral-1b-base reaches up to 233.5 tok/s and averages 230.4 tok/s, with the fastest results on NVIDIA GeForce RTX 3060 Laptop GPU.
- How much memory does ministral-1b-base need?
- The leanest observed configuration used about 2.7 GB of memory.
- Which tools have been used to run ministral-1b-base?
- Benchmarks were submitted using Bionic. Results are community-contributed and updated as new runs arrive.