llm-benchmarks
githubdrose.io

Qwen4 Max Benchmarks


Measured on alibaba

26 runs
Avg Tokens / Second
19.20
Decode Tokens / Second
40.30
Avg Time to First Token (ms)
2030.00
Runs Analysed
26
Last Updated
Nov 22, 2026, 07:45 AM

Throughput distribution

Throughput over time

Qwen4 Maxalibaba19tok/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 · models with at least 4 samples in the window first, fastest of those at the top · shared vertical scale, 0 to 28 tok/s (99th percentile) · dashed rule is the model's own mean

Every lane measuring this model

2 measured · live first · a lane is a provider, a catalogue id and a transport
ProviderLaneAvg Toks/SecMinMaxAvg TTF (ms)n
alibabavia OpenRouter19.209.2826.302030.0026
alibabadirect17.8017.8017.801870.001

Frequently Asked Questions

How fast is Qwen4 Max?
The latest rolling average throughput is 19.20 tokens per second with an average time to first token of 2030.00 ms across 26 recent runs. Once tokens are flowing the decoder delivers about 40.30 tokens per second, measured from the gaps between stream chunks with time to first token and reasoning excluded.
How often are these benchmarks updated?
Benchmarks refresh automatically whenever the monitoring cron runs. The most recent run completed on Nov 22, 2026, 07:45 AM.

Related models