llm-benchmarks
githubdrose.io

MiniMax M3 Benchmarks


Measured on coreweave

110 runs
Avg Tokens / Second
23.30
Avg Time to First Token (ms)
3030.00
Runs Analysed
110
Last Updated
Sep 1, 2026, 12:49 PM

Throughput distribution

Throughput over time

MiniMax M3coreweave3tok/s

Generated throughput, including reasoning tokens — so a thinking model's line does not collapse; the table ranks on visible tokens · one cell per model, showing its best-covered provider · shared vertical scale, 0 to 143 tok/s (99th percentile) · dashed rule is the model's own mean · 14 further providers not drawn

Every provider serving this model

2 measured
ProviderModelAvg Toks/SecMinMaxAvg TTF (ms)
coreweaveMiniMax M328.209.8561.102140.00
coreweaveMiniMax M318.406.9324.603920.00

Frequently Asked Questions

How fast is MiniMax M3?
The latest rolling average throughput is 23.30 tokens per second with an average time to first token of 3030.00 ms across 110 recent runs.
How often are these benchmarks updated?
Benchmarks refresh automatically whenever the monitoring cron runs. The most recent run completed on Sep 1, 2026, 12:49 PM.

Related models