llm-benchmarks
githubdrose.io

Qwen3 VL 8B Thinking Benchmarks


Measured on alibaba

40 runs
Avg Tokens / Second
15.15
Avg Time to First Token (ms)
4275.00
Runs Analysed
40
Last Updated
Sep 1, 2026, 04:07 AM

Throughput distribution

Throughput over time

Qwen3 VL 8B Thinkingalibaba3tok/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 129 tok/s (99th percentile) · dashed rule is the model's own mean · 1 further provider not drawn

Every provider serving this model

2 measured
ProviderModelAvg Toks/SecMinMaxAvg TTF (ms)
alibabaQwen3 VL 8B Thinking14.1012.2015.804150.00
alibabaQwen3 VL 8B Thinking16.2010.10113.004400.00

Frequently Asked Questions

How fast is Qwen3 VL 8B Thinking?
The latest rolling average throughput is 15.15 tokens per second with an average time to first token of 4275.00 ms across 40 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, 04:07 AM.

Related models