Epoch ECI leaderboard: which LLMs score highest

Epoch ECI (Epoch Capabilities Index: Epoch's own aggregate of ~40 benchmarks) currently has scores for 205 models. The top score is 157 by Qwen 3.8 Max; the median model scores 127 and the lowest scores 56.

Epoch ECI leaderboard: top 50 of 205 models (highest-effort row per model)
#ModelOrganisationEpoch ECI scoreCheapest input $/MSourceCaptured
1Qwen 3.8 MaxAlibaba157—Epoch AI Benchmarking Hub2026-10-05
2Gemini 3.1 ProGoogle DeepMind155$2.00Epoch AI Benchmarking Hub2026-10-05
3Qwen3.7 MaxAlibaba154$1.25Epoch AI Benchmarking Hub2026-10-05
4Gemini 3 ProGoogle DeepMind153$2.00Epoch AI Benchmarking Hub2026-10-05
5Muse SparkMeta AI152—Epoch AI Benchmarking Hub2026-10-05
6grok-4-20xAI152$1.25Epoch AI Benchmarking Hub2026-10-05
7Gemini 3 FlashGoogle DeepMind152$0.500Epoch AI Benchmarking Hub2026-10-05
8Kimi K2.6Moonshot151$0.650Epoch AI Benchmarking Hub2026-10-05
9GPT-5 ProOpenAI150$15.00Epoch AI Benchmarking Hub2026-10-05
10Kimi K2.7 CodeMoonshot150$0.671Epoch AI Benchmarking Hub2026-10-05
11GLM-5.1Z.ai (Zhipu AI)150$1.05Epoch AI Benchmarking Hub2026-10-05
12Qwen 3.8 27BAlibaba149$0.990Epoch AI Benchmarking Hub2026-10-05
13Qwen 3.6 Max (Preview)Alibaba149—Epoch AI Benchmarking Hub2026-10-05
14Grok 4.3 BetaxAI149—Epoch AI Benchmarking Hub2026-10-05
15Kimi K2.5Moonshot148$0.450Epoch AI Benchmarking Hub2026-10-05
16Qwen 3.6 PlusAlibaba148—Epoch AI Benchmarking Hub2026-10-05
17Qwen3.7 PlusAlibaba147$0.282Epoch AI Benchmarking Hub2026-10-05
18MiniMax-M3MiniMax147$0.230Epoch AI Benchmarking Hub2026-10-05
19Qwen 3.5 Plus (hosted 397B-A17B)Alibaba147—Epoch AI Benchmarking Hub2026-10-05
20MiniMax-M2.5MiniMax147$0.270Epoch AI Benchmarking Hub2026-10-05
21Qwen3.5 397B-A17BAlibaba147$0.164Epoch AI Benchmarking Hub2026-10-05
22Qwen3.6 27BAlibaba146$0.150Epoch AI Benchmarking Hub2026-10-05
23Grok 4xAI146$3.00Epoch AI Benchmarking Hub2026-10-05
24DeepSeek-V3.2DeepSeek146$0.259Epoch AI Benchmarking Hub2026-10-05
25nemotron-3-ultraNvidia146$0.500Epoch AI Benchmarking Hub2026-10-05
26Kimi K2 ThinkingMoonshot146$0.550Epoch AI Benchmarking Hub2026-10-05
27MiniMax-M2.7MiniMax146$0.210Epoch AI Benchmarking Hub2026-10-05
28GLM-5Z.ai (Zhipu AI)146$0.600Epoch AI Benchmarking Hub2026-10-05
29Gemini 2.5 Pro (Jun 2025)Google DeepMind145$1.25Epoch AI Benchmarking Hub2026-10-05
30Qwen3.7 FlashAlibaba145$0.030Epoch AI Benchmarking Hub2026-10-05
31grok-4-fastxAI144—Epoch AI Benchmarking Hub2026-10-05
32Gemini 2.5 Pro (Mar 2025)Google DeepMind144$1.25Epoch AI Benchmarking Hub2026-10-05
33Qwen 3.5 Flash (hosted 35B-A3B)Alibaba144—Epoch AI Benchmarking Hub2026-10-05
34Qwen 3.6 35B-A3BAlibaba144—Epoch AI Benchmarking Hub2026-10-05
35Qwen3-235B-A22B-Thinking (Jul 2025)Alibaba144$0.180Epoch AI Benchmarking Hub2026-10-05
36GLM-4.7Z.ai (Zhipu AI)144$0.400Epoch AI Benchmarking Hub2026-10-05
37Qwen 3.6 FlashAlibaba143—Epoch AI Benchmarking Hub2026-10-05
38Gemini 2.5 Flash (Sep 2025)Google DeepMind143$0.300Epoch AI Benchmarking Hub2026-10-05
39Gemma 4 31B ITGoogle DeepMind143$0.100Epoch AI Benchmarking Hub2026-10-05
40Claude Opus 4Anthropic143$15.00Epoch AI Benchmarking Hub2026-10-05
41qwen3.5-35b-a3bAlibaba143$0.140Epoch AI Benchmarking Hub2026-10-05
42GPT-5.5 InstantOpenAI142—Epoch AI Benchmarking Hub2026-10-05
43Gemini 2.5 Pro (May 2025)Google DeepMind142$1.25Epoch AI Benchmarking Hub2026-10-05
44Qwen3-MaxAlibaba142$0.780Epoch AI Benchmarking Hub2026-10-05
45Gemma 4 26B A4BGoogle DeepMind142$0.076Epoch AI Benchmarking Hub2026-10-05
46Gemini 2.5 Flash (May 2025)Google DeepMind142$0.300Epoch AI Benchmarking Hub2026-10-05
47Mistral Medium 3.5Mistral AI141$1.50Epoch AI Benchmarking Hub2026-10-05
48DeepSeek-R1 (May 2025)DeepSeek141$0.400Epoch AI Benchmarking Hub2026-10-05
49Gemini 2.5 Flash (Jun 2025)Google DeepMind141$0.300Epoch AI Benchmarking Hub2026-10-05
50Grok 3 minixAI140$0.250Epoch AI Benchmarking Hub2026-10-05

Compare all benchmarks side by side on the live leaderboard →

Among the ten highest scorers, the cheapest listed API price belongs to Gemini 3 Flash at $0.500 per million input tokens (score 152).

What Epoch ECI measures

Epoch ECI measures Epoch Capabilities Index: Epoch's own aggregate of ~40 benchmarks.

How to read these scores

Higher scores are better; the scale is specific to this benchmark, so compare models only on this board. Where a model is listed at several reasoning efforts, only its highest-effort row is shown. See the methodology for how rows are chosen.

Where the data comes from

Epoch AI Benchmarking Hub (CC BY 4.0). Scores were last captured 2026-10-05; the Captured column gives each row's date.

Related benchmarks