How many tracked prints does the OmniScience Non-Hallucination family have?

OmniScience Non-Hallucination leaderboard — the family board.

2 TRACKED VARIANTS · 1 FIRST-CLASS BOARD

One family, 2 tracked prints. Scoring configs and splits of the base board render as sections of this page, each under its own honest label.

● LIVE DATA355 MODELS ON THIS BOARDLEADER MINICPM5-1B AT 99.1

The board — OmniScience Non-HallucinationTOP 100 OF 355 MEASURED · BEST SCORE PER MODEL, DEFAULT VARIANT

#ModelVendorScoreVLast seen

01MiniCPM5-1BOpenBMB99.12026-06-05

02G9v3-3BAI9Stars88.32026-07-23

03G9v3-39A5BAI9Stars87.02026-08-04

04Command A+Cohere85.82026-06-19

05Gemini 4 ArgonGoogle84.92026-10-01

06LFM2.5-2.6BLiquid AI84.02026-08-22

07Grok 4.3SpaceXAI83.12026-06-01

08Qwen3.8 27BAlibaba81.92026-08-25

09MiniMax M3MiniMax81.62026-06-04

10MiniCPM5-2BOpenBMB78.12026-09-08

11Ling-3.0-flash-VLInclusionAI78.02026-09-11

12Grok 4.20SpaceXAI77.62026-09-10

13K-EXAONE 2.0LG AI77.42026-08-29

14Grok 4.6SpaceXAI76.02026-08-22

15Solar Pro4Upstage75.62026-08-13

16MiMo V2.5 ProXiaomi75.32026-09-22

17Solar Open2 250BUpstage74.62026-08-13

18Granite 4.2 30BIBM74.42026-08-28

19Qwen3.7 MaxAlibaba74.42026-06-01

20Claude Haiku 4.5Anthropic74.32026-07-20

21Grok 3 Mini ReasoningSpaceXAI74.22026-06-01

22K2 Horizon 375B A23BMBZUAI74.12026-09-04

23GLM-5.2Z.ai73.72026-07-11

24Granite 4.2 3BIBM73.72026-08-27

25GLM 5.3 FlashZ.ai72.42026-08-27

26Qwen3.7 Plus PreviewAlibaba72.32026-06-04

27Motif-3Motif Technologies71.62026-08-13

28Qwen3.8 Max PreviewAlibaba71.22026-09-16

29Claude Sonnet 4Anthropic70.92026-06-23

30Grok 4.7SpaceXAI70.72026-10-02

31GLM-5.3Z.ai70.42026-09-25

32Nemotron 3 Ultra 550B A55BNVIDIA70.32026-06-06

33GLM-5.1Z.ai70.12026-06-01

34MiMo V2 ProXiaomi70.02026-06-01

35Gemma 3 270MGoogle69.62026-06-01

36Ling 3.0 TinyInclusionAI69.52026-08-07

37K2 Horizon 36B-A4BMBZUAI69.52026-09-15

38Gemma 4 E4BGoogle69.12026-06-01

39Muse Spark 1.3Meta68.52026-09-03

40MiMo V2.5Xiaomi68.12026-09-24

41Granite 4.2 8BIBM68.02026-08-27

42Gemma 4 E2BGoogle67.62026-06-01

43Agnes 2.5 Pro BetaSapiens AI67.22026-08-27

44A.X-K2SK Telecom67.02026-08-13

45Muse Spark 1.2Meta66.72026-08-06

46K2 Horizon 7BMBZUAI66.42026-09-15

47Gemini 3.5 Flash LiteGoogle65.62026-07-22

48Qwen3.6 PlusAlibaba65.42026-06-01

49GLM-5Z.ai64.72026-06-01

50MiniMax M2.7MiniMax64.42026-06-01

51Solar Mini4Upstage64.22026-10-01

52Qwen3.5 Omni PlusAlibaba63.02026-06-01

53Nemotron 3.5 LightningNVIDIA62.42026-08-12

54Ling-3.1-flashInclusionAI62.12026-10-04

55GPT-4oOpenAI62.12026-09-10

56Qwen3.8 2.4T A95BAlibaba60.82026-08-14

57Claude Opus 4.8Anthropic60.72026-10-04

58Claude Sonnet 5Anthropic60.62026-10-04

59Ling-3.0-flash-FinInclusionAI60.52026-09-17

60Claude 3.7 SonnetAnthropic60.02026-06-01

61MiMo V2 OmniXiaomi59.92026-06-01

62Haiku 5.5Anthropic59.62026-10-08

63Kimi K2.6Moonshot59.52026-06-01

64MiMo V2.6 ProXiaomi59.42026-09-22

65Claude 3.5 HaikuAnthropic58.92026-06-01

66Mistral Large 4Mistral58.12026-10-07

67Claude Opus 4.7Anthropic57.72026-10-04

68Step 5 PreviewStepFun57.02026-09-19

69Llama 3.1 8B InstructMeta57.02026-06-01

70JT-4.1 Flash 236B A21BChina Mobile56.32026-09-30

71Ling-3.0-flashInclusionAI55.92026-08-06

72K2 Horizon 3.7BMBZUAI55.92026-09-15

73GPT-5 miniOpenAI55.72026-06-21

74GPT-6 AstraOpenAI55.22026-09-04

75Qwen3 Coder 480B A35B InstructAlibaba55.12026-06-01

76Qwen3.8 Flash NextAlibaba54.72026-08-28

77Qwen3.6 Max PreviewAlibaba53.82026-06-01

78LFM2.5-8B-A1BLiquid AI53.12026-06-01

79Claude Sonnet 5.5 (Adaptive Reasoning, Max Effort, Default Fallback)Anthropic53.02026-10-04

80Agnes 3.0 FlashSapiens AI51.62026-09-12

81MiMo V2 FlashXiaomi51.62026-06-01

82Claude Sonnet 4.6Anthropic51.62026-10-04

83Claude Sonnet 4.5Anthropic50.92026-06-01

84Motif-3-BetaMotif Technologies50.92026-07-21

85Qwen3.6 27BAlibaba50.72026-06-01

86GPT-6.1 SolOpenAI50.62026-09-30

87Kimi K2.5Moonshot50.02026-06-01

88Muse Spark 1.1Meta50.02026-07-11

89Claude Sonnet 5.5 (Adaptive Reasoning, Low Effort, Default Fallback)Anthropic49.82026-10-04

90Qwen3.6 35B A3BAlibaba49.52026-06-01

91GPT-6 SolOpenAI49.32026-09-23

92Gemini 3.1 ProGoogle49.12026-06-01

93Kimi K3Moonshot49.12026-07-31

94GPT-5.4 nanoOpenAI48.92026-09-10

95Claude 5.5Anthropic48.82026-10-04

96GLM-4.6VZ.ai48.52026-06-01

97GPT-5.1 Codex miniOpenAI48.42026-06-01

98GPT-5.1OpenAI48.12026-06-20

99Gemini 3.8 FlashGoogle48.12026-09-03

100Llama 3.1 Instruct 405BMeta47.62026-06-21

◆ IV — INDEPENDENTLY VERIFIED · ◆ AA — AGGREGATOR · ◆ VA — VENDOR · ◇ SA — SOURCE-ATTRIBUTED

Variants of this boardSCORING CONFIGS & SPLITS — NOT CITED STANDALONE, SO THEY LIVE HERE

OmniScience Accuracy

VARIANT354 models · best score per model

#ModelVendorScoreVLast seen

01Claude Fable 5.1Anthropic67.22026-10-04

02Claude Opus 5.5Anthropic66.22026-10-04

03Claude Fable 5Anthropic65.32026-10-04

04Claude Opus 5.5 (Adaptive Reasoning, Medium Effort, Default Fallback)Anthropic64.52026-10-04

05GPT-6 AstraOpenAI62.62026-09-04

06GPT-6.1 SolOpenAI62.12026-09-30

07Claude Opus 5Anthropic60.92026-10-04

08GPT-5.6 SolOpenAI59.42026-07-10

09GPT-5.5OpenAI58.02026-06-01

10Gemini 3 ProGoogle55.82026-06-01

11Gemini 3.7 FlashGoogle55.32026-08-14

12Gemini 3.1 ProGoogle54.92026-06-01

13Gemini 3.8 FlashGoogle54.62026-09-03

14GPT-6 SolOpenAI54.52026-09-23

15Claude Sonnet 5.5 (Adaptive Reasoning, Max Effort, Default Fallback)Anthropic54.02026-10-04

THE FAMILY BOARD · EVERY PRINT LABELED · EVERY SCORE SOURCE-ATTRIBUTED ON ITS OWN LIVE BOARDAll leaderboards The frontier board Models by vendor