InferenceBench

Qwen/Qwen2.5-7B-Instruct

vllm unknown on 8x NVIDIA H100 80GB HBM3 (bf16) · seed 0 · run 019ebfa5-665c-7445-8796-ef49aceb5316 · signed

Headline metrics

TTFT P50 (ms) 16.43
TTFT P99 (ms)
Throughput (tok/s)
$/M tokens
J/token 6.58
Power avg (W) 869
Power peak (W) 938
WER mean
J / audio s

All metrics

energy_joules_total 7,127
joules_per_token 6.58
n_ok 5.00
n_samples 5.00
ok_rate 1.00
pass_at_1 1.00
pass_at_1_p05 1.00
pass_at_1_p50 1.00
pass_at_1_p95 1.00
power_avg_w 869
power_peak_w 938
timeout_rate 0.0000
tokens_out_total 1,083
total_p50_ms 1,406
ttft_p50_ms 16.43

Provenance

Model revisionunknown00
Providervllm
Hardware fingerprintc353425077b20b36c71d9c32d9e5d49096478f4a0f7d6078c1dfc83c408ccf50
Datasetbuiltin-mbpp-mini · 9dd0eb0ef0d05cee08d36b6485ebf2855f24e2ecb863cd704393a0edbd38d1a9
Timestamp2026-06-13T06:22:28.444424+00:00
Envelope JSON111a690bba4e.json

Verify this result

bench verify /bench/leaderboard/envelopes/111a690bba4e.json

bench verify re-downloads this envelope, recomputes the canonical content hash, and validates the Sigstore signature against the embedded certificate + Rekor entry.