CtrlAltDebrief

Omniscience benchmark

303 AI models ranked on Omniscience. Each row shows the model’s score next to its cheapest API price per 1M tokens where it has one, so you can weigh quality against cost.

Top model today

Claude Fable 5 (Anthropic) leads at 40.1, from $10.00 per 1M input.

Data via arXiv:2511.13029 →

Models ranked

303

on this benchmark

Top score

40.1

current leader

Type

Single eval

benchmark

Updated

15 July 2026

last source fetch

Cost vs Omniscience

161 priced models
Most attractive quadrant (cheap + high)
= reasoning model (extended thinking)
-85-59-33-71945$0.03$0.1$0.3$1$3$10COST / 1M INPUT TOKENS (log) →OMNISCIENCE SCORE ↑Claude Fable 5Gemini 3.1 Pro PreviewClaude Opus 4.8Grok 4.5Claude Opus 4.7Gemini 3.5 FlashGPT-5.6 SolGPT-5.5Grok 4.3GPT-5.1MiMo-V2.5-ProQwen3.6 PlusMiniMax M3MiniMax M2.7Grok 3 MiniGPT-5 MiniGPT-5.4 NanoQwen3.6 27BQwen3.6 35B A3BGemma 4 E2bLlama 3.1 8B InstructLlama 3.2 1B Instructgpt-oss-20b

Each labelled name links to that model’s page. Curated to the cost-efficiency frontier, top scorers and cheapest so labels stay readable; the full field is in the leaderboard table below (and via “select all”). Reasoning models (extended “thinking” before answering) carry a ring around the dot.

Models ranked on Omniscience

highest score first · 303 models
Maker
#
1Claude Fable 5Anthropic40.1$10.00$50.00
2Gemini 3.1 Pro PreviewGoogle32.9$2.00$12.00
3Claude Opus 4.8Anthropic27.4$5.00$25.00
4Grok 4.5xAI26.4$2.00$6.00
5Claude Opus 4.7Anthropic26.2$5.00$25.00
6Gemini 3.5 FlashGoogle22.7$1.50$9.00
7GPT-5.6 SolOpenAI21.7$5.00$30.00
8GPT-5.5OpenAI20.1$5.00$30.00
9Grok 4.3xAI18.3$1.25$2.50
10Muse Spark 1.1Other18$1.25$4.25
11Gemini 3 ProGoogle15.8$2.00$12.00
12Grok 4.20 0309 v2Spacexai15.3n/an/a
13Claude Sonnet 5Anthropic15.3$2.00$10.00
14Qwen3.7 MaxAlibaba14.1n/an/a
15Claude Opus 4.6Anthropic13.5reasoning: adaptive$5.00$25.00
16Grok 4.20 0309Spacexai13.4n/an/a
17Claude Opus 4.5Anthropic13.3reasoning: true$5.00$25.00
18Claude Sonnet 4.6Anthropic12.4reasoning: adaptive$3.00$15.00
19Gemini 3 FlashGoogle11.6reasoning: truen/an/a
20GPT-5.5 Instant (May 2026)OpenAI10.3n/an/a
21Qwen3.6 Max PreviewAlibaba10.2n/an/a
22GPT-5.3 CodexOpenAI9.9n/an/a
23Grok Build 0.1 0616Spacexai7.4n/an/a
24Kimi K2.6Moonshot6.4n/an/a
25GPT-5.4OpenAI5.7$2.50$15.00
26GPT-5.1OpenAI5.5$1.25$10.00
27MiMo-V2-ProXiaomi4.9n/an/a
28Muse SparkMeta4.1n/an/a
29GLM 5.2Zhipu4.0$1.40$4.40
30Grok 4xAI3.8$3.00$15.00
31MiMo-V2.5-ProXiaomi3.6$1.00$3.00
32GPT-5.5 Instant (June 2026)OpenAI3.5n/an/a
33Qwen3.6 PlusAlibaba2.6$0.33$1.95
34Qwen3.7 PlusAlibaba2.4n/an/a
35GLM 5Zhipu2$0.80$2.56
36GLM 5.1Zhipu1.9$1.05$3.50
37MiniMax M3MiniMax1.4$0.30$1.20
38MiniMax M2.7MiniMax0.7$0.26$0.55
39Claude 4.5 SonnetAnthropic0.3reasoning: true$3.00$15.00
40Claude 3.7 SonnetAnthropic0.3reasoning: true$3.00$15.00
41Claude 4 SonnetAnthropic0.2reasoning: true$3.00$15.00
42GPT-5.2OpenAI0.0medium$1.75$14.00
43GPT-5.6 TerraOpenAI-0.2$2.50$15.00
44MiniCPM5-1BOpenbmb-0.8reasoning: falsen/an/a
45Nemotron 3 UltraNVIDIA-0.8n/an/a
46GPT-5.2-CodexOpenAI-2.5$1.75$14.00
47Command A+Cohere-4.0n/an/a
48Claude 4.5 HaikuAnthropic-4.2reasoning: true$1.00$5.00
49GPT-5.1 CodexOpenAI-6.0n/an/a
50Grok 3 MinixAI-6.0reasoning: true$0.25$0.50
1–50 of 303
Page 1 of 7
Prices are on-demand list rates. Enterprise commitments can be materially lower. Cost per task uses best-effort token usage.

Frequently asked questions

What is the Omniscience benchmark?

AA-Omniscience (Artificial Analysis) measures factual-knowledge reliability and calibration across 6,000 questions in six domains. Its Omniscience Index rewards correct answers and appropriate abstentions while penalising confident wrong answers, so a model that guesses badly can score below zero.

Which AI model scores highest on Omniscience?

Claude Fable 5 (Anthropic) leads with 40.1, from $10.00 per 1M input tokens.

How many models are ranked on Omniscience?

303 models carry a Omniscience score. Scores come from Artificial Analysis; where a model has a live API price it is verified against LiteLLM and OpenRouter, and models with no current price are listed without one.

Other benchmarks

compare the same models on a different eval
Intelligence Indexcomposite indexCoding Indexcomposite indexAgentic Indexcomposite indexGPQA DiamondleaderboardHumanity's Last ExamleaderboardMMLU-ProleaderboardMMMU-ProleaderboardLiveCodeBenchleaderboardSciCodeleaderboardTerminalBench HardleaderboardTerminalBench v2.1leaderboardτ²-benchleaderboardτ-bench BankingleaderboardAPEX AgentsleaderboardIT-Bench SREleaderboardGDPvalleaderboardCritPtleaderboardIFBenchleaderboardLong-Context Reasoningcomposite indexSWE-bench VerifiedleaderboardSWE-bench ProleaderboardAIME 2026leaderboardARC-AGI-2leaderboardAA-Briefcaseleaderboard
Benchmark scores via Artificial Analysis · pricing verified against LiteLLM + OpenRouter · refreshed daily · updated 15 July 2026

Every weekday

AI moves fast. Here's your debrief.

News, analysis, tools, and more.

For people who build with AI