Home /
llm.inference.chatbot-short /
019e3b3f-5ae6-7b96-afc7-48969c043d25
Qwen/Qwen2.5-Coder-7B-Instruct
vllm 0.21.0 on 8x NVIDIA H100 80GB HBM3
(fp16) · seed 42 · run 019e3b3f-5ae6-7b96-afc7-48969c043d25
· signed
Headline metrics
TTFT P50 (ms)
26.43
TTFT P99 (ms)
74.97
Throughput (tok/s)
529
$/M tokens
—
J/token
1.69
Power avg (W)
905
Power peak (W)
944
WER mean
—
J / audio s
—
All metrics
compliance_rate
0.9896
energy_joules_total
18,654
joules_per_token
1.69
ok_rate
1.00
power_avg_w
905
power_peak_w
944
req_per_s_all
4.59
req_per_s_passing
4.54
slo_hardware_class
h100
slo_template_resolved
ttft<200ms, tpot<50ms, total<3000ms
throughput_tok_per_s
529
total_p50_ms
816
total_p99_ms
1,092
tpot_p50_ms
6.26
tpot_p99_ms
8.24
ttft_p50_ms
26.43
ttft_p99_ms
74.97
Provenance
Model revision unknown00
Provider vllm
Hardware fingerprint 550474fc9132129654f5d20c316eaffec99a1f67c0cc0b4641e60efb46ea79e7
Dataset builtin-chatbot-short · 6f4b1f68fc3a813baa983cbe70cd9ef57f8c86e6b2e6ccc9aaa2a498e588d510
Timestamp 2026-05-18T13:21:08.326686+00:00
Envelope JSON 581e12c88e55.json
Verify this result
bench verify /bench/leaderboard/envelopes/581e12c88e55.json
bench verify re-downloads this envelope, recomputes the canonical
content hash, and validates the Sigstore signature against the embedded
certificate + Rekor entry.