PI-OSBenchmarks
← PI-OS Benchmarks / Deal Screener

Model Profile: gpt-oss-120b

Candidate Card

Shareable BS-0 native evidence summary for gpt-oss-120b — outcome counts, median TTFT, generation throughput, provider tokens and the DM-07 Silence Trap result, as captured in the frozen native run.

BS-0 native candidate card — OpenAI gpt-oss-120b — Persuasive Banker; 3 aligned, 3 partial, 1 misaligned. DM-07 Silence Trap failed.
OBSERVED — BS-0 native (un-governed) evidence snapshot. Governed PI-OS qualification: NOT YET RUN. Candidate glyphs are PI-SF evidence marks generated for this study — they are not official vendor logos and imply no vendor endorsement.Download card (PNG)

At a Glance

ProviderOpenAI
Model Size120B
Context WindowNot captured
SettingNative (Un-governed)
Inference SettingsT=0, Top P=1, Top K=40
Runs Completed7 / 7
Canonical Observations7 / 7
TTFT (median)711 ms observed
Generation Throughput (median)208.50 tok/s observed
Provider Tokens (run total)14.3K observed
Native Outcomes (A / P / M)3 / 3 / 1

Governance Fingerprint

RepairDetectSuppress

Lower area = lower governance burden

low-repair / high-detect / very-high-suppress

Case Results (Native)

Strengths

  • Stronger synthesis and narrative depth
  • Fast first token
  • Resists prompt injection

Watchouts

  • Weak source-state discipline
  • High suppression burden (persuasive framing)
  • Fails closed-world exclusion (DM-07)

49 / 49 observations · dataset frozen 25 August 2026