Examenos

68.6% · raw 68.6 %

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: AraGen v3 (3C3H)09-2025.

79.7% · raw 79.7 %

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: Correctness.

78.3% · raw 78.3 %

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: Completeness.

24.4% · raw 24.4 %

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: Conciseness.

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: Helpfulness.

77.9% · raw 77.9 %

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: Honesty.

78.8% · raw 78.8 %

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: Harmlessness.

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: Arabic MMLU*Accuracy. Audited subset: items with invalid options, ambiguous questions or indefensible gold labels removed.

94.4% · raw 94.4 %

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: BelebeleAccuracy.

94.7% · raw 94.7 %

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: AraSTEM*Accuracy. Audited subset: items with invalid options, ambiguous questions or indefensible gold labels removed.

92.2% · raw 92.2 %

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 GSM8K*Accuracy. Audited subset: items with invalid options, ambiguous questions or indefensible gold labels removed.

97.5% · raw 97.5 %

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: GATmath*Accuracy. Audited subset: items with invalid options, ambiguous questions or indefensible gold labels removed.

85.4% · raw 85.4 %

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: MedAraBench*Accuracy. Audited subset: items with invalid options, ambiguous questions or indefensible gold labels removed.

69.3% · raw 69.3 %

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.

89.7% · raw 89.7 %

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: IslamicQAAccuracy (mean of the 4 below).

93.4% · raw 93.4 %

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: PaLM-X 2025-2*Accuracy. Audited subset: items with invalid options, ambiguous questions or indefensible gold labels removed.

95.7% · raw 95.7 %

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-1*Accuracy. Audited subset: items with invalid options, ambiguous questions or indefensible gold labels removed.

87.2% · raw 87.2 %

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: QIAS-2*Accuracy. Audited subset: items with invalid options, ambiguous questions or indefensible gold labels removed.

82.7% · raw 82.7 %

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: IslamicEval2025-1B*Accuracy. Audited subset: items with invalid options, ambiguous questions or indefensible gold labels removed.

87.5% · raw 87.5 %

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: Arab CultureAccuracy (mean of the 3 below).

88.6% · raw 88.6 %

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: ArabCulture*Accuracy. Audited subset: items with invalid options, ambiguous questions or indefensible gold labels removed.

85.5% · raw 85.5 %

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: AraDiCE-Culture*Accuracy. Audited subset: items with invalid options, ambiguous questions or indefensible gold labels removed.

88.5% · raw 88.5 %

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: Arabic Cuisine^Accuracy. Vendor says benchmark to be released; no public test-set availability assumed.

87.3% · raw 87.3 %

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: Dialectal Arabic MMLU*Accuracy. Audited subset: items with invalid options, ambiguous questions or indefensible gold labels removed.

69.1% · raw 69.1 %

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.

83% · raw 83 %

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.

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 MMLU with diacriticsAccuracy.

71.5% · raw 71.5 %

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: FannOrFlop-MCQAccuracy.

94% · raw 94 %

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.

90.7% · raw 90.7 %

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: DiacritizationAccuracy.

97.3% · raw 97.3 %

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: ParsingAccuracy.

90.8% · raw 90.8 %

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: FLORES200+LLM-as-a-judge.

90.8% · raw 90.8 %

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.

92% · raw 92 %

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.

80% · raw 80 %

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.

73% · raw 73 %

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.

73% · raw 73 %

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.

66% · raw 66 %

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.

31.5% · 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: 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.

5.7% · raw 5.7 %

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: F-score.

83.1% · raw 83.1 %

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: Non-hallucination rate.

67.3% · raw 67.3 %

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: Terminal Bench 2.1Accuracy (Avg@3). Accuracy Avg@3.

67.3% · raw 67.3 %

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.

67.7% · raw 67.7 %

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: GDPValAccuracy (@ 75% threshold). Task accuracy at a 75% rubric threshold, not Elo or mean rubric completion.

67.7% · raw 67.7 %

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.

69.3% · raw 69.3 %

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: MCPMarkAccuracy (Pass@1).

69.3% · raw 69.3 %

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.

40.3% · raw 40.3 %

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: WildClawBench (English text-only subset)Mean Score@3. English text-only subset, Mean Score@3; not full multimodal suite or pass@3.

40.3% · raw 40.3 %

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.

23.6% · raw 23.6 %

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.

23.6% · raw 23.6 %

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.

51.5% · raw 51.5 %

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: GAIA v2 lit (Ar)Accuracy (Pass@1).

51.5% · raw 51.5 %

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.

52.3% · raw 52.3 %

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 File ProcessingAccuracy (@ 75% threshold). Task accuracy at a 75% rubric threshold, not Elo or mean rubric completion.

52.3% · raw 52.3 %

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.

72.8% · raw 72.8 %

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: SWE-Bench Verified (Offline)Accuracy (Pass@1). Offline protocol.

72.8% · raw 72.8 %

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.

71.8% · raw 71.8 %

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: BrowseComp-PlusAccuracy.

71.8% · raw 71.8 %

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.

82.5% · raw 82.5 %

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.

88.7% · raw 88.7 %

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.

91% · raw 91 %

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.

91.1% · raw 91.1 %

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.

94.8% · raw 94.8 %

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.

94.8% · raw 94.8 %

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.

77.3% · raw 77.3 %

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% · raw 85 %

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.

88.1% · raw 88.1 %

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.

88.1% · raw 88.1 %

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.

79.8% · raw 79.8 %

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.

79.8% · raw 79.8 %

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.

84.1% · raw 84.1 %

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.

68.5% · raw 68.5 %

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.

86% · raw 86 %

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.

48.8% · raw 48.8 %

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.

57.4% · raw 57.4 %

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.