THE SCOREBOARD FOR WHAT'S NEXTAI + QUANTUM, SCORED. COMPARED. EXPLAINED.
AIQUANTUMSCORE.

THE EVIDENCE / AGENTS

Agents
benchmarks.

Completion of multi-step tool-using tasks. Results are grouped by benchmark and unit. Different tests are never merged into a raw leaderboard.

01 / RESULTS

Published results

BenchmarkVersionMethodConfigurationModelResultTest dateVerifiedSource typeSource
APEX-AgentsAPEX-AgentsPublisher-reported configuration; see source—grok-4-657.5 %2026-08-122026-09-29provider first partySource ↗
APEX-AgentsAPEX-AgentsPublisher-reported configuration; see source—grok-4-547.1 %2026-08-122026-09-29provider first partySource ↗
Terminal-Bench v3.0Terminal-Bench v3.0Publisher-reported configuration; see source—grok-4-626 %2026-08-122026-09-29provider first partySource ↗
Terminal-Bench v3.0Terminal-Bench v3.0Publisher-reported configuration; see source—grok-4-515.7 %2026-08-122026-09-29provider first partySource ↗
GDPVal-AA v2GDPVal-AA v2Publisher-reported configuration; see source—grok-4-61753 Elo2026-08-122026-09-29provider first partySource ↗
GDPVal-AA v2GDPVal-AA v2Publisher-reported configuration; see source—grok-4-51526 Elo2026-08-122026-09-29provider first partySource ↗
GDPval-AA v2.1 (reported best effort)GDPval-AA v2.1 (reported best effort)Publisher-reported configuration; see source—claude-sonnet-5-51844 Elo2026-09-282026-09-29provider first partySource ↗
GDPval-AA v2.1 (reported best effort)GDPval-AA v2.1 (reported best effort)Publisher-reported configuration; see source—claude-opus-5-51846 Elo2026-09-282026-09-29provider first partySource ↗
GDPval-AA v2.1 (reported best effort)GDPval-AA v2.1 (reported best effort)Publisher-reported configuration; see source—gpt-6-sol1487 Elo2026-09-282026-09-29provider first partySource ↗
AA-Briefcase v1.1 (reported best effort)AA-Briefcase v1.1 (reported best effort)Publisher-reported configuration; see source—claude-sonnet-5-51811 Elo2026-09-282026-09-29provider first partySource ↗
AA-Briefcase v1.1 (reported best effort)AA-Briefcase v1.1 (reported best effort)Publisher-reported configuration; see source—claude-opus-5-51822 Elo2026-09-282026-09-29provider first partySource ↗
AA-Briefcase v1.1 (reported best effort)AA-Briefcase v1.1 (reported best effort)Publisher-reported configuration; see source—gpt-6-sol1483 Elo2026-09-282026-09-29provider first partySource ↗
OSWorld 2.1 (partial; reported best effort)OSWorld 2.1 (partial; reported best effort)Publisher-reported configuration; see source—claude-sonnet-5-580.1 %2026-09-282026-09-29provider first partySource ↗
OSWorld 2.1 (partial; reported best effort)OSWorld 2.1 (partial; reported best effort)Publisher-reported configuration; see source—claude-opus-5-581.8 %2026-09-282026-09-29provider first partySource ↗
SkillsBench Vals OpenHands with skillsVals 2026-09-27Vals OpenHands harness with skills—deepseek-v4-1-flash69.8 %2026-09-272026-09-29independent labSource ↗
SkillsBench Vals OpenHands with skillsVals 2026-09-27Vals OpenHands harness with skills—grok-4-566.03 %2026-09-272026-09-29independent labSource ↗
SkillsBench Vals OpenHands with skillsVals 2026-09-27Vals OpenHands harness with skills—gemini-3-7-flash65.89 %2026-09-272026-09-29independent labSource ↗
AutomationBench-AA Index v4.3.2 (Opus Max fallback, Astra Max, Gemini High)AA Index 4.3.2Opus Max with default fallback; Astra Max—claude-opus-5-570 %2026-09-292026-09-29independent labSource ↗
AutomationBench-AA Index v4.3.2 (Opus Max fallback, Astra Max, Gemini High)AA Index 4.3.2Opus Max with default fallback; Astra Max—gpt-6-astra68 %2026-09-292026-09-29independent labSource ↗
AutomationBench-AA Index v4.3.2 (Opus Max fallback, Astra Max, Gemini High)AA Index 4.3.2Gemini High—gemini-3-8-flash60 %2026-09-292026-09-29independent labSource ↗
LiveBench javascript2026-06-25Public table; published model effort variantLiveBench public table 2026-06-25 · gemini-3.1-pro-preview-highgemini-3-1-pro-preview59.091 %not published2026-09-30benchmark maintainerSource ↗
LiveBench javascript2026-06-25Public table; published model effort variantLiveBench public table 2026-06-25 · gemini-3.5-flash-highgemini-3-5-flash63.636 %not published2026-09-30benchmark maintainerSource ↗
LiveBench javascript2026-06-25Public table; published model effort variantLiveBench public table 2026-06-25 · grok-4.5grok-4-572.727 %not published2026-10-17benchmark maintainerSource ↗
LiveBench javascript2026-06-25Public table; published model effort variantLiveBench public table 2026-06-25 · gpt-5.6-sol-maxgpt-5.6-sol63.636 %not published2026-10-17benchmark maintainerSource ↗
LiveBench javascript2026-06-25Public table; published model effort variantLiveBench public table 2026-06-25 · gpt-5.6-terra-maxgpt-5.6-terra68.182 %not published2026-10-17benchmark maintainerSource ↗
LiveBench javascript2026-06-25Public table; published model effort variantLiveBench public table 2026-06-25 · gpt-5.6-luna-maxgpt-5.6-luna63.636 %not published2026-10-17benchmark maintainerSource ↗
LiveBench javascript2026-06-25Public table; published model effort variantLiveBench public table 2026-06-25 · gemini-3.5-flash-lite-highgemini-3-5-flash-lite59.091 %not published2026-09-30benchmark maintainerSource ↗
LiveBench javascript2026-06-25Public table; published model effort variantLiveBench public table 2026-06-25 · gemini-3.6-flash-highgemini-3-6-flash63.636 %not published2026-09-30benchmark maintainerSource ↗
LiveBench javascript2026-06-25Public table; published model effort variantLiveBench public table 2026-06-25 · deepseek-v4-pro-0813deepseek-v4-pro68.182 %not published2026-09-30benchmark maintainerSource ↗
LiveBench javascript2026-06-25Public table; published model effort variantLiveBench public table 2026-06-25 · grok-4.6grok-4-672.727 %not published2026-10-17benchmark maintainerSource ↗
LiveBench javascript2026-06-25Public table; published model effort variantLiveBench public table 2026-06-25 · gemini-3.7-flash-highgemini-3-7-flash68.182 %not published2026-10-17benchmark maintainerSource ↗
LiveBench javascript2026-06-25Public table; published model effort variantLiveBench public table 2026-06-25 · claude-fable-5-1-max-effortclaude-fable-5-168.182 %not published2026-10-17benchmark maintainerSource ↗
LiveBench javascript2026-06-25Public table; published model effort variantLiveBench public table 2026-06-25 · gemini-3.8-flash-highgemini-3-8-flash72.727 %not published2026-10-17benchmark maintainerSource ↗
LiveBench javascript2026-06-25Public table; published model effort variantLiveBench public table 2026-06-25 · gpt-6-astra-maxgpt-6-astra63.636 %not published2026-10-17benchmark maintainerSource ↗
LiveBench javascript2026-06-25Public table; published model effort variantLiveBench public table 2026-06-25 · deepseek-v4.1-flash-maxdeepseek-v4-1-flash81.818 %not published2026-10-17benchmark maintainerSource ↗
LiveBench javascript2026-06-25Public table; published model effort variantLiveBench public table 2026-06-25 · claude-opus-5-5-max-effortclaude-opus-5-581.818 %not published2026-10-17benchmark maintainerSource ↗
LiveBench javascript2026-06-25Public table; published model effort variantLiveBench public table 2026-06-25 · gpt-6-sol-maxgpt-6-sol63.636 %not published2026-10-17benchmark maintainerSource ↗
LiveBench javascript2026-06-25Public table; published model effort variantLiveBench public table 2026-06-25 · gpt-6-luna-maxgpt-6-luna63.636 %not published2026-10-17benchmark maintainerSource ↗
LiveBench javascript2026-06-25Public table; published model effort variantLiveBench public table 2026-06-25 · claude-sonnet-5-5-max-effortclaude-sonnet-5-577.273 %not published2026-10-17benchmark maintainerSource ↗
AutomationBench-AA AA October frontier cohortAA Index 4.3.2Artificial Analysis independent evaluation; published reasoning configuration per modelAA Index 4.3.2 October frontier comparison; first-party models; effort recorded separately · Highgemini-4-argon78 %not published2026-10-16independent labSource ↗
AutomationBench-AA AA October frontier cohortAA Index 4.3.2Artificial Analysis independent evaluation; published reasoning configuration per modelAA Index 4.3.2 October frontier comparison; first-party models; effort recorded separately · Maxmuse-wonder68 %not published2026-10-16independent labSource ↗
LiveBench 2.0 long_agentic20261025LiveBench 2.0 published release tablePublished model configuration; effort not specified in tableclaude-fable-5-177.9 %not published2026-11-07benchmark maintainerSource ↗
LiveBench 2.0 long_agentic20261025LiveBench 2.0 published release tablePublished model configuration; effort not specified in tableclaude-opus-5-572.8 %not published2026-11-07benchmark maintainerSource ↗
LiveBench 2.0 long_agentic20261025LiveBench 2.0 published release tablePublished model configuration; effort not specified in tableclaude-sonnet-5-563.4 %not published2026-11-07benchmark maintainerSource ↗
LiveBench 2.0 long_agentic20261025LiveBench 2.0 published release tablePublished model configuration; effort not specified in tablegemini-3-8-flash16.5 %not published2026-11-07benchmark maintainerSource ↗
LiveBench 2.0 long_agentic20261025LiveBench 2.0 published release tablePublished model configuration; effort not specified in tablegemini-4-argon61.2 %not published2026-11-07benchmark maintainerSource ↗
LiveBench 2.0 long_agentic20261025LiveBench 2.0 published release tablePublished model configuration; effort not specified in tablegpt-5.6-terra33.1 %not published2026-11-07benchmark maintainerSource ↗
LiveBench 2.0 long_agentic20261025LiveBench 2.0 published release tablePublished model configuration; effort not specified in tablegpt-6-astra77.5 %not published2026-11-07benchmark maintainerSource ↗
LiveBench 2.0 long_agentic20261025LiveBench 2.0 published release tablePublished model configuration; effort not specified in tablegpt-6-luna30.9 %not published2026-11-07benchmark maintainerSource ↗
LiveBench 2.0 long_agentic20261025LiveBench 2.0 published release tablePublished model configuration; effort not specified in tablegpt-6.1-sol68.7 %not published2026-11-07benchmark maintainerSource ↗
LiveBench 2.0 long_agentic20261025LiveBench 2.0 published release tablePublished model configuration; effort not specified in tablegrok-4-628 %not published2026-11-07benchmark maintainerSource ↗

Limitations: Tool sets and time budgets differ across evaluations. Publication methods, prompts, sampling and model versions may differ. Check each cited source before interpreting a result.