VERIFIED INTELLIGENCE
Changes link to their source. Unclear monitor candidates stay in the review queue until validated. Check update status →
Last seven days: 2026-09-25 through 2026-10-01. See all verified changes →
AIQuantumScore’s derived coding score changed under methodology 3.0.0. Peer results and evidence freshness can affect this score; it is not a new benchmark claim by a provider.
OpenAI · derived from verified evidence · verified 2026-10-01 · 72 → 63 derived coding score
Inspect source ↗Historical snapshot from 2026-10-01, not a live score. Current coding score for GPT-6 Astra: 74. Scores can change again as comparable evidence is added.
gpt-6-astra ↗AIQuantumScore’s derived coding score changed under methodology 3.0.0. Peer results and evidence freshness can affect this score; it is not a new benchmark claim by a provider.
OpenAI · derived from verified evidence · verified 2026-10-01 · 0 → 39 derived coding score
Inspect source ↗Historical snapshot from 2026-10-01, not a live score. Current coding score for GPT-6 Sol: 47. Scores can change again as comparable evidence is added.
gpt-6-sol ↗AIQuantumScore’s derived coding score changed under methodology 3.0.0. Peer results and evidence freshness can affect this score; it is not a new benchmark claim by a provider.
Anthropic · derived from verified evidence · verified 2026-10-01 · 8 → 43 derived coding score
Inspect source ↗Historical snapshot from 2026-10-01, not a live score. Current coding score for Claude Fable 5.1: 62. Scores can change again as comparable evidence is added.
claude-fable-5-1 ↗AIQuantumScore’s derived coding score changed under methodology 3.0.0. Peer results and evidence freshness can affect this score; it is not a new benchmark claim by a provider.
Anthropic · derived from verified evidence · verified 2026-10-01 · 82 → 87 derived coding score
Inspect source ↗Historical snapshot from 2026-10-01, not a live score. Current coding score for Claude Opus 5.5: 94. Scores can change again as comparable evidence is added.
claude-opus-5-5 ↗