Examenos

72.4% · raw 72.4 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: AraGen v3 (3C3H)09-2025.

82.1% · raw 82.1 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: Correctness.

80.9% · raw 80.9 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: Completeness.

31.5% · raw 31.5 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: Conciseness.

78.2% · raw 78.2 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: Helpfulness.

80.7% · raw 80.7 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: Honesty.

81.1% · raw 81.1 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: Harmlessness.

89.3% · raw 89.3 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: Arabic MMLU*Accuracy. Audited subset: items with invalid options, ambiguous questions or indefensible gold labels removed.

95.6% · raw 95.6 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: BelebeleAccuracy.

95.5% · raw 95.5 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: AraSTEM*Accuracy. Audited subset: items with invalid options, ambiguous questions or indefensible gold labels removed.

86.8% · raw 86.8 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: Arabic GSM8K*Accuracy. Audited subset: items with invalid options, ambiguous questions or indefensible gold labels removed.

98.4% · raw 98.4 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: GATmath*Accuracy. Audited subset: items with invalid options, ambiguous questions or indefensible gold labels removed.

89.5% · raw 89.5 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: MedAraBench*Accuracy. Audited subset: items with invalid options, ambiguous questions or indefensible gold labels removed.

71.3% · raw 71.3 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: Arabic Legal (HELM)*Accuracy. Audited subset: items with invalid options, ambiguous questions or indefensible gold labels removed.

90.2% · raw 90.2 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: IslamicQAAccuracy (mean of the 4 below).

95.3% · raw 95.3 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: PaLM-X 2025-2*Accuracy. Audited subset: items with invalid options, ambiguous questions or indefensible gold labels removed.

95.4% · raw 95.4 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: QIAS-1*Accuracy. Audited subset: items with invalid options, ambiguous questions or indefensible gold labels removed.

87.7% · raw 87.7 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: QIAS-2*Accuracy. Audited subset: items with invalid options, ambiguous questions or indefensible gold labels removed.

82.3% · raw 82.3 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: IslamicEval2025-1B*Accuracy. Audited subset: items with invalid options, ambiguous questions or indefensible gold labels removed.

89.1% · raw 89.1 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: Arab CultureAccuracy (mean of the 3 below).

89.8% · raw 89.8 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: ArabCulture*Accuracy. Audited subset: items with invalid options, ambiguous questions or indefensible gold labels removed.

87.7% · raw 87.7 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: AraDiCE-Culture*Accuracy. Audited subset: items with invalid options, ambiguous questions or indefensible gold labels removed.

89.8% · raw 89.8 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: Arabic Cuisine^Accuracy. Vendor says benchmark to be released; no public test-set availability assumed.

89.6% · raw 89.6 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: Dialectal Arabic MMLU*Accuracy. Audited subset: items with invalid options, ambiguous questions or indefensible gold labels removed.

71.1% · raw 71.1 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: Dialect Identification (QADI)Accuracy.

87.6% · raw 87.6 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: Alyah*Accuracy. Audited subset: items with invalid options, ambiguous questions or indefensible gold labels removed.

88.2% · raw 88.2 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: Arabic MMLU with diacriticsAccuracy.

73.2% · raw 73.2 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: FannOrFlop-MCQAccuracy.

94.8% · raw 94.8 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: Arabic Linguistics^Accuracy (mean of the 2 below). Vendor says benchmark to be released; no public test-set availability assumed.

91.9% · raw 91.9 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: DiacritizationAccuracy.

97.7% · raw 97.7 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: ParsingAccuracy.

91.9% · raw 91.9 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: FLORES200+LLM-as-a-judge.

91.9% · raw 91.9 %

Alternative · high effort · Own vendor

Source/record date: 2026-10-17

As reported by IFM/Inception42. Final-checkpoint launch summary; Arabic overall/domain averages weight benchmark components, not displayed subdomain averages. Agentic average covers nine suites, BrowseComp replaces Plus at flagship scale; those averages are not identical test pools. Card definitions preserve audited sets/thresholds/subsets.

93.9% · raw 93.9 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: Arabic IFBench^Accuracy (prompt-level strict). Vendor says benchmark to be released; no public test-set availability assumed.

77.7% · raw 77.7 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: Arabic IFEvalAccuracy.

77% · raw 77 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: AA-LCR (Ar)Accuracy.

77% · raw 77 %

Alternative · high effort · Own vendor

Source/record date: 2026-10-17

As reported by IFM/Inception42. Final-checkpoint launch summary; Arabic overall/domain averages weight benchmark components, not displayed subdomain averages. Agentic average covers nine suites, BrowseComp replaces Plus at flagship scale; those averages are not identical test pools. Card definitions preserve audited sets/thresholds/subsets.

70.9% · raw 70.9 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: HalluScoreAccuracy.

31.2% · raw 31.2 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: SimpleQA Verified (Ar)Aggregate (2/3 F-score + 1/3 non-hallucination rate). Aggregate = two-thirds F-score + one-third non-hallucination rate; not ordinary accuracy.

8.5% · raw 8.5 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: F-score.

76.7% · raw 76.7 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: Non-hallucination rate.

73.8% · raw 73.8 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: Terminal Bench 2.1Accuracy (Avg@3). Accuracy Avg@3.

73.8% · raw 73.8 %

Alternative · high effort · Own vendor

Source/record date: 2026-10-17

As reported by IFM/Inception42. Final-checkpoint launch summary; Arabic overall/domain averages weight benchmark components, not displayed subdomain averages. Agentic average covers nine suites, BrowseComp replaces Plus at flagship scale; those averages are not identical test pools. Card definitions preserve audited sets/thresholds/subsets.

73.2% · raw 73.2 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: GDPValAccuracy (@ 75% threshold). Task accuracy at a 75% rubric threshold, not Elo or mean rubric completion.

73.2% · raw 73.2 %

Alternative · high effort · Own vendor

Source/record date: 2026-10-17

As reported by IFM/Inception42. Final-checkpoint launch summary; Arabic overall/domain averages weight benchmark components, not displayed subdomain averages. Agentic average covers nine suites, BrowseComp replaces Plus at flagship scale; those averages are not identical test pools. Card definitions preserve audited sets/thresholds/subsets.

68.5% · raw 68.5 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: MCPMarkAccuracy (Pass@1).

68.5% · raw 68.5 %

Alternative · high effort · Own vendor

Source/record date: 2026-10-17

As reported by IFM/Inception42. Final-checkpoint launch summary; Arabic overall/domain averages weight benchmark components, not displayed subdomain averages. Agentic average covers nine suites, BrowseComp replaces Plus at flagship scale; those averages are not identical test pools. Card definitions preserve audited sets/thresholds/subsets.

50.4% · raw 50.4 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: WildClawBench (English text-only subset)Mean Score@3. English text-only subset, Mean Score@3; not full multimodal suite or pass@3.

50.4% · raw 50.4 %

Alternative · high effort · Own vendor

Source/record date: 2026-10-17

As reported by IFM/Inception42. Final-checkpoint launch summary; Arabic overall/domain averages weight benchmark components, not displayed subdomain averages. Agentic average covers nine suites, BrowseComp replaces Plus at flagship scale; those averages are not identical test pools. Card definitions preserve audited sets/thresholds/subsets.

18.6% · raw 18.6 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: Apex-Agents (420 text-only subset, loop agent)Pass@1 (Accuracy). 420-task text-only subset, loop agent.

18.6% · raw 18.6 %

Alternative · high effort · Own vendor

Source/record date: 2026-10-17

As reported by IFM/Inception42. Final-checkpoint launch summary; Arabic overall/domain averages weight benchmark components, not displayed subdomain averages. Agentic average covers nine suites, BrowseComp replaces Plus at flagship scale; those averages are not identical test pools. Card definitions preserve audited sets/thresholds/subsets.

50.3% · raw 50.3 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: GAIA v2 lit (Ar)Accuracy (Pass@1).

50.3% · raw 50.3 %

Alternative · high effort · Own vendor

Source/record date: 2026-10-17

As reported by IFM/Inception42. Final-checkpoint launch summary; Arabic overall/domain averages weight benchmark components, not displayed subdomain averages. Agentic average covers nine suites, BrowseComp replaces Plus at flagship scale; those averages are not identical test pools. Card definitions preserve audited sets/thresholds/subsets.

64.1% · raw 64.1 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: Arabic File ProcessingAccuracy (@ 75% threshold). Task accuracy at a 75% rubric threshold, not Elo or mean rubric completion.

64.1% · raw 64.1 %

Alternative · high effort · Own vendor

Source/record date: 2026-10-17

As reported by IFM/Inception42. Final-checkpoint launch summary; Arabic overall/domain averages weight benchmark components, not displayed subdomain averages. Agentic average covers nine suites, BrowseComp replaces Plus at flagship scale; those averages are not identical test pools. Card definitions preserve audited sets/thresholds/subsets.

70.2% · raw 70.2 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: SWE-Bench Verified (Offline)Accuracy (Pass@1). Offline protocol.

70.2% · raw 70.2 %

Alternative · high effort · Own vendor

Source/record date: 2026-10-17

As reported by IFM/Inception42. Final-checkpoint launch summary; Arabic overall/domain averages weight benchmark components, not displayed subdomain averages. Agentic average covers nine suites, BrowseComp replaces Plus at flagship scale; those averages are not identical test pools. Card definitions preserve audited sets/thresholds/subsets.

72.6% · raw 72.6 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

Harness: IFM/Inception42 final main checkpoint, high effort. IFM/Inception42 final-checkpoint evaluation. High effort. Arabic suites use generation-based evaluation; starred source tests use audited cleaned question sets, not the unfiltered public suite. Exact source label: BrowseCompDiscard-all@95k. Discard-all context management at 95K.

72.6% · raw 72.6 %

Alternative · high effort · Own vendor

Source/record date: 2026-10-17

As reported by IFM/Inception42. Final-checkpoint launch summary; Arabic overall/domain averages weight benchmark components, not displayed subdomain averages. Agentic average covers nine suites, BrowseComp replaces Plus at flagship scale; those averages are not identical test pools. Card definitions preserve audited sets/thresholds/subsets.

84% · raw 84 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

As reported by IFM/Inception42. Final-checkpoint launch summary; Arabic overall/domain averages weight benchmark components, not displayed subdomain averages. Agentic average covers nine suites, BrowseComp replaces Plus at flagship scale; those averages are not identical test pools. Card definitions preserve audited sets/thresholds/subsets.

89.5% · raw 89.5 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

As reported by IFM/Inception42. Final-checkpoint launch summary; Arabic overall/domain averages weight benchmark components, not displayed subdomain averages. Agentic average covers nine suites, BrowseComp replaces Plus at flagship scale; those averages are not identical test pools. Card definitions preserve audited sets/thresholds/subsets.

92.5% · raw 92.5 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

As reported by IFM/Inception42. Final-checkpoint launch summary; Arabic overall/domain averages weight benchmark components, not displayed subdomain averages. Agentic average covers nine suites, BrowseComp replaces Plus at flagship scale; those averages are not identical test pools. Card definitions preserve audited sets/thresholds/subsets.

92.4% · raw 92.4 %

Alternative · high effort · Own vendor

Source/record date: 2026-10-17

Figure-only summary printed on official model card; differs in some rounded values from launch blog. Blog summary remains headline; no averaging or reconstruction.

93.5% · raw 93.5 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

As reported by IFM/Inception42. Final-checkpoint launch summary; Arabic overall/domain averages weight benchmark components, not displayed subdomain averages. Agentic average covers nine suites, BrowseComp replaces Plus at flagship scale; those averages are not identical test pools. Card definitions preserve audited sets/thresholds/subsets.

93.6% · raw 93.6 %

Alternative · high effort · Own vendor

Source/record date: 2026-10-17

Figure-only summary printed on official model card; differs in some rounded values from launch blog. Blog summary remains headline; no averaging or reconstruction.

80.4% · raw 80.4 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

As reported by IFM/Inception42. Final-checkpoint launch summary; Arabic overall/domain averages weight benchmark components, not displayed subdomain averages. Agentic average covers nine suites, BrowseComp replaces Plus at flagship scale; those averages are not identical test pools. Card definitions preserve audited sets/thresholds/subsets.

86.7% · raw 86.7 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

As reported by IFM/Inception42. Final-checkpoint launch summary; Arabic overall/domain averages weight benchmark components, not displayed subdomain averages. Agentic average covers nine suites, BrowseComp replaces Plus at flagship scale; those averages are not identical test pools. Card definitions preserve audited sets/thresholds/subsets.

89.4% · raw 89.4 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

As reported by IFM/Inception42. Final-checkpoint launch summary; Arabic overall/domain averages weight benchmark components, not displayed subdomain averages. Agentic average covers nine suites, BrowseComp replaces Plus at flagship scale; those averages are not identical test pools. Card definitions preserve audited sets/thresholds/subsets.

89.4% · raw 89.4 %

Alternative · high effort · Own vendor

Source/record date: 2026-10-17

Figure-only summary printed on official model card; differs in some rounded values from launch blog. Blog summary remains headline; no averaging or reconstruction.

82.8% · raw 82.8 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

As reported by IFM/Inception42. Final-checkpoint launch summary; Arabic overall/domain averages weight benchmark components, not displayed subdomain averages. Agentic average covers nine suites, BrowseComp replaces Plus at flagship scale; those averages are not identical test pools. Card definitions preserve audited sets/thresholds/subsets.

82.8% · raw 82.8 %

Alternative · high effort · Own vendor

Source/record date: 2026-10-17

Figure-only summary printed on official model card; differs in some rounded values from launch blog. Blog summary remains headline; no averaging or reconstruction.

85.4% · raw 85.4 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

As reported by IFM/Inception42. Final-checkpoint launch summary; Arabic overall/domain averages weight benchmark components, not displayed subdomain averages. Agentic average covers nine suites, BrowseComp replaces Plus at flagship scale; those averages are not identical test pools. Card definitions preserve audited sets/thresholds/subsets.

70.1% · raw 70.1 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

As reported by IFM/Inception42. Final-checkpoint launch summary; Arabic overall/domain averages weight benchmark components, not displayed subdomain averages. Agentic average covers nine suites, BrowseComp replaces Plus at flagship scale; those averages are not identical test pools. Card definitions preserve audited sets/thresholds/subsets.

85.8% · raw 85.8 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

As reported by IFM/Inception42. Final-checkpoint launch summary; Arabic overall/domain averages weight benchmark components, not displayed subdomain averages. Agentic average covers nine suites, BrowseComp replaces Plus at flagship scale; those averages are not identical test pools. Card definitions preserve audited sets/thresholds/subsets.

51% · raw 51 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

As reported by IFM/Inception42. Final-checkpoint launch summary; Arabic overall/domain averages weight benchmark components, not displayed subdomain averages. Agentic average covers nine suites, BrowseComp replaces Plus at flagship scale; those averages are not identical test pools. Card definitions preserve audited sets/thresholds/subsets.

60.2% · raw 60.2 %

Headline · high effort · Own vendor

Source/record date: 2026-10-17

As reported by IFM/Inception42. Final-checkpoint launch summary; Arabic overall/domain averages weight benchmark components, not displayed subdomain averages. Agentic average covers nine suites, BrowseComp replaces Plus at flagship scale; those averages are not identical test pools. Card definitions preserve audited sets/thresholds/subsets.