Model intelligence

34 AI models

Capability (Epoch Capabilities Index and individual benchmarks from Epoch AI), API list prices and context windows (OpenRouter), release dates and open-weight status. Benchmarks are shown individually - never merged into one invented score.

Leads on ECI

Claude Opus 5.5

ECI 167.35Epoch AI

Leads on GPQA Diamond

GPT-6 Astra

95.8%Epoch AI

Leads on SWE-bench Verified

Claude Opus 4.7

83.5%Epoch AI

Cheapest in ECI top 10

Claude Sonnet 5.5

$2.00/M inputOpenRouter

Longest context

Grok 4.20 Multi-Agent

2M tokensOpenRouter

Best open-weight on ECI

Kimi K3

ECI 157.61Epoch AI

Epoch Capabilities Index by release date

Capability frontier

About ECI ↗

Source: Epoch AI (CC BY 4.0). ECI is Epoch's item-response model over many benchmarks; each model page shows its 90% confidence interval.

Blended $/1M tokens (3:1 input:output) vs ECI

Price vs capability

Only models with both an ECI score and an OpenRouter price. Normalisation: 75% input + 25% output price. Click a point to open the model.

34 models

1 filter:

#	Model	ECI ↓	AA idx	GPQA◆	SWE-v	FrontierMath	AIME	HLE	Terminal	Context	In $/M	Out $/M	Weights	Released
1	Muse Spark 1.3 Meta· reasoning	156.9	—	—	—	74.0	—	—	—	1.05M	$1.25	$4.25	Closed	2 Sep 2026
2	Muse Spark 1.2 Meta· reasoning	155.0	39.6	—	—	—	—	—	—	1.05M	$1.25	$4.25	Closed	5 Aug 2026
3	Muse Spark 1.1 Meta· reasoning	154.3	33.7	—	—	—	—	—	—	1.05M	$1.25	$4.25	Closed	9 Jul 2026
4	Muse Spark Meta	152.1	—	89.8	—	—	88.9	40.6	—	—	—	—	Closed	8 Apr 2026
5	Llama 4 Maverick Meta	132.2	—	—	—	—	—	5.7	—	1.05M	$0.19	$0.65	Open source	6 Apr 2025
6	Llama 4 Scout Meta	129.7	—	51.8	—	—	7.8	—	—	1.31M	$0.10	$0.30	Open source	5 Apr 2025
7	Llama 3.1-405B Meta	128.8	—	50.9	—	—	9.7	—	—	—	—	—	Open source	23 Jul 2024
8	Llama 3.3 70B Meta	127.3	—	47.4	—	—	5.1	—	—	—	—	—	Open source	6 Dec 2024
9	Llama 3.1-70B Meta	125.9	—	44.2	—	—	3.6	—	—	—	—	—	Open source	23 Jul 2024
10	Llama 3.2 90B Meta	125.5	—	41.0	—	—	2.6	—	—	—	—	—	Open source	24 Sep 2024
11	Llama 3-70B Meta	122.9	—	40.6	—	—	4.3	—	—	—	—	—	Open source	18 Apr 2024
12	Llama 3.1-8B Meta	116.6	—	27.0	—	—	1.7	—	—	—	—	—	Open source	23 Jul 2024
13	Llama 3-8B Meta	116.5	—	26.1	—	—	1.9	—	—	—	—	—	Open source	18 Apr 2024
14	Llama 2-70B Meta	113.8	—	26.3	—	—	0.0	—	—	—	—	—	Open source	18 Jul 2023
15	LLaMA-65B Meta	110.2	—	—	—	—	—	—	—	—	—	—	Open source	24 Feb 2023
16	LLaMA-33B Meta	107.4	—	—	—	—	—	—	—	—	—	—	Open source	24 Feb 2023
17	Llama 2-13B Meta	106.2	—	—	—	—	—	—	—	—	—	—	Open source	18 Jul 2023
18	Llama 2-34B Meta	105.2	—	—	—	—	—	—	—	—	—	—	Closed	18 Jul 2023
19	Llama 3.2 1B Meta	102.0	—	23.9	—	—	0.6	—	—	—	—	—	Open source	24 Sep 2024
20	LLaMA-13B Meta	100.6	—	—	—	—	—	—	—	—	—	—	Open source	24 Feb 2023
21	Llama 2-7B Meta	99.1	—	—	—	—	—	—	—	—	—	—	Open source	18 Jul 2023
22	LLaMA-7B Meta	96.7	—	—	—	—	—	—	—	—	—	—	Open source	24 Feb 2023
23	Muse Spark 1.3 Contributor Meta· reasoning	—	—	—	—	—	—	—	—	1.05M	$0.10	$0.20	?	2 Sep 2026
24	Muse Spark 1.2 Contributor Meta· reasoning	—	—	—	—	—	—	—	—	1.05M	$0.10	$0.20	?	21 Aug 2026
25	Muse Glimmer 30B Meta· reasoning	—	—	—	—	—	—	—	—	131K	$0.35	$1.50	Open source	9 Aug 2026
26	Llama Guard 4 12B Meta	—	—	—	—	—	—	—	—	164K	$0.18	$0.18	Open source	30 Apr 2025
27	Llama 4 Maverick (FP8) Meta	—	—	67.0	—	—	20.6	—	—	—	—	—	Open source	5 Apr 2025
28	Llama 3.3 70B Instruct Meta	—	—	—	—	—	—	—	—	131K	$0.10	$0.32	Open source	6 Dec 2024
29	Llama 3.2 1B Instruct Meta	—	—	—	—	—	—	—	—	60K	$0.027	$0.20	Open source	25 Sep 2024
30	Llama 3.2 3B Instruct Meta	—	—	—	—	—	—	—	—	131K	$0.050	$0.33	Open source	25 Sep 2024
31	Llama 3.2 11B Meta	—	—	—	—	—	—	—	—	—	—	—	Open source	24 Sep 2024
32	Llama 3.2 3B Meta	—	—	—	—	—	—	—	—	—	—	—	Open source	24 Sep 2024
33	Llama 3.1 8B Instruct Meta	—	—	—	—	—	—	—	—	131K	$0.050	$0.080	Open source	23 Jul 2024
34	Llama 3.1 70B Instruct Meta	—	—	—	—	—	—	—	—	131K	$0.40	$0.40	Open source	23 Jul 2024

Showing 34 of 34. Benchmark cells are % (best reported setting). 3.9K benchmark results in total.Export CSV