Model rankings

Rank	Change	Model	Score	95% interval	Votes	Price $/M
1	—	llama-3.1-405b-instruct-bf16OpenMeta	1,252	±7	11,826	—
2	—	llama-3.1-405b-instruct-fp8OpenMeta	1,251	±4	36,694	—
3	—	mistral-large-2407OpenMistral	1,230	±3	32,970	—
4	▲ 1	athene-70b-0725OpenNexusFlow	1,220	±4	20,689	—
5	▼ 1	qwen2.5-72b-instructOpenAlibaba	1,218	±6	11,306	—
6	—	deepseek-v2.5OpenDeepSeek	1,217	±6	11,330	—
7	—	llama-3.1-70b-instructOpenMeta	1,211	±4	31,908	—
8	—	gemma-2-27b-itOpenGoogle	1,203	±4	43,220	—
9	—	jamba-1.5-largeOpenAI21 Labs	1,202	±7	9,179	—
10	—	gemma-2-9b-it-simpoOpenPrinceton	1,197	±5	10,612	—
11	—	nemotron-4-340b-instructOpenNVIDIA	1,195	±5	20,641	—
12	—	llama-3-70b-instructOpenMeta	1,194	±3	163,833	—
13	—	command-r-plus-08-2024OpenCohere	1,191	±7	9,115	—
14	—	command-r-plusOpenCohere	1,182	±3	80,862	—
15	▲ 1	deepseek-coder-v2OpenDeepSeek	1,181	±5	15,791	—
16	▼ 1	qwen2-72b-instructOpenAlibaba	1,180	±3	38,973	—
17	—	gemma-2-9b-itOpenGoogle	1,175	±4	33,147	—
18	—	command-r-08-2024OpenCohere	1,174	±7	9,460	—
19	—	jamba-1.5-miniOpenAI21 Labs	1,158	±6	9,294	—
20	—	qwen1.5-110b-chatOpenAlibaba	1,155	±4	27,442	—
21	—	mixtral-8x22b-instruct-v0.1OpenMistral	1,150	±3	53,817	—
22	▲ 1	command-rOpenCohere	1,148	±3	56,369	—
23	▼ 1	qwen1.5-72b-chatOpenAlibaba	1,147	±4	40,610	—
24	—	llama-3-8b-instructOpenMeta	1,141	±3	109,318	—
25	—	yi-1.5-34b-chatOpen01 AI	1,131	±4	25,141	—
26	—	zephyr-orpo-141b-A35b-v0.1OpenHuggingFace	1,131	±7	4,858	—
27	—	qwen1.5-32b-chatOpenAlibaba	1,126	±4	22,738	—
28	—	llama-3.1-8b-instructOpenMeta	1,124	±5	30,110	—
29	—	phi-3-medium-4k-instructOpenMicrosoft	1,118	±4	26,141	—
30	▲ 16	internlm2_5-20b-chatOpenInternLM	1,115	±9	5,076	—
31	▼ 1	mixtral-8x7b-instruct-v0.1OpenMistral	1,114	±0	76,106	—
32	▼ 1	dbrx-instruct-previewOpenDatabricks	1,114	±4	33,704	—
33	—	gemma-2-2b-itOpenGoogle	1,109	±5	24,658	—
34	▼ 2	qwen1.5-14b-chatOpenAlibaba	1,106	±4	18,656	—
35	▼ 1	wizardlm-70bOpenMicrosoft	1,105	±7	8,383	—
36	▼ 1	snowflake-arctic-instructOpenSnowflake	1,103	±3	34,158	—
37	▼ 1	openchat-3.5-0106OpenOpenChat · 1 more variant	1,101	±4	12,964	—
38	▼ 1	yi-34b-chatOpen01 AI	1,101	±5	15,918	—
39	—	gemma-1.1-7b-itOpenGoogle	1,098	±4	25,056	—
40	▼ 2	phi-3-small-8k-instructOpenMicrosoft	1,098	±4	18,487	—
41	—	tulu-2-dpo-70bOpenAllenAI/UW	1,093	±8	6,662	—
42	—	vicuna-33bOpenLMSYS	1,091	±5	22,929	—
43	▲ 1	deepseek-llm-67b-chatOpenDeepSeek	1,091	±8	4,979	—
44	▲ 1	openhermes-2.5-mistral-7bOpenNousResearch	1,091	±9	5,090	—
45	▼ 2	starling-lm-7b-betaOpenNexusflow	1,091	±5	16,660	—
46	▼ 6	llama-3.2-3b-instructOpenMeta	1,090	±9	8,604	—
47	—	starling-lm-7b-alphaOpenUC Berkeley	1,083	±7	10,413	—
48	—	llama-2-70b-chatOpenMeta	1,081	±4	39,612	—
49	—	llama2-70b-steerlm-chatOpenNVIDIA	1,074	±11	3,635	—
50	—	nous-hermes-2-mixtral-8x7b-dpoOpenNousResearch	1,072	±11	3,839	—
51	—	phi-3-mini-4k-instruct-june-2024OpenMicrosoft	1,069	±6	12,877	—
52	—	mistral-7b-instruct-v0.2OpenMistral	1,067	±4	20,050	—
53	—	qwen1.5-7b-chatOpenAlibaba	1,066	±10	4,861	—
54	▲ 2	solar-10.7b-instruct-v1.0OpenUpstage AI	1,066	±9	4,286	—
55	▼ 1	wizardlm-13bOpenMicrosoft	1,064	±8	7,183	—
56	▲ 1	dolphin-2.2.1-mistral-7bOpenCognitive Computations	1,063	±12	1,714	—
57	▼ 2	falcon-180b-chatOpenTII	1,061	±16	1,327	—
58	—	mpt-30b-chatOpenMosaicML	1,060	±13	2,649	—
59	—	vicuna-13bOpenLMSYS	1,060	±5	19,785	—
60	▲ 1	qwen-14b-chatOpenAlibaba	1,057	±9	5,068	—
61	▲ 1	phi-3-mini-4k-instructOpenMicrosoft	1,055	±4	21,116	—
62	▲ 1	llama-2-13b-chatOpenMeta	1,053	±4	19,736	—
63	▲ 1	zephyr-7b-betaOpenHuggingFace	1,050	±6	11,321	—
64	▲ 1	phi-3-mini-128k-instructOpenMicrosoft	1,049	±5	21,589	—
65	▲ 1	gemma-7b-itOpenGoogle	1,047	±6	9,176	—
66	▲ 1	codellama-34b-instructOpenMeta	1,047	±6	7,504	—
67	▲ 2	zephyr-7b-alphaOpenHuggingFace	1,041	±13	1,814	—
68	▼ 8	llama-3.2-1b-instructOpenMeta	1,040	±11	8,699	—
69	▼ 1	guanaco-33bOpenUW	1,040	±16	2,996	—
70	—	gemma-1.1-2b-itOpenGoogle	1,039	±6	11,352	—
71	▲ 1	stripedhyena-nous-7bOpenTogether AI	1,035	±7	5,269	—
72	▼ 1	codellama-70b-instructOpenMeta	1,034	±16	1,189	—
73	—	mistral-7b-instructOpenMistral	1,027	±6	9,137	—
74	—	llama-2-7b-chatOpenMeta	1,023	±5	14,548	—
75	—	vicuna-7bOpenLMSYS	1,022	±7	7,018	—
76	—	qwen1.5-4b-chatOpenAlibaba	1,014	±7	7,805	—
77	—	gemma-2b-itOpenGoogle	1,011	±11	4,915	—
78	—	olmo-7b-instructOpenAi2	993	±9	6,488	—
79	—	gpt4all-13b-snoozyOpenNomic AI	983	±21	1,786	—
80	—	koala-13bOpenUC Berkeley	982	±10	7,033	—
81	—	chatglm3-6bOpenTsinghua	976	±10	4,764	—
82	—	alpaca-13bOpenStanford	970	±11	5,870	—
83	—	mpt-7b-chatOpenMosaicML	969	±10	4,015	—
84	—	RWKV-4-Raven-14BOpenRWKV	951	±12	4,933	—
85	▲ 1	chatglm2-6bOpenTsinghua	936	±12	2,707	—
86	▼ 1	oasst-pythia-12bOpenOpenAssistant	933	±11	6,379	—
87	—	fastchat-t5-3bOpenLMSYS	900	±11	4,300	—
88	—	chatglm-6bOpenTsinghua	891	±12	4,994	—
89	—	dolly-v2-12bOpenDatabricks	887	±13	3,485	—
90	—	llama-13bOpenMeta	874	±16	2,444	—
91	—	stablelm-tuned-alpha-7bOpenStability AI	864	±13	3,339	—

41 more models

Intelligence over time

Each maker's best model at the time on the Artificial Analysis Intelligence Index, by the month it was released.

Gained most · Anthropic +57.2Highest now · Anthropic 64.6Gap between highest and lowest 18.1 ptsTime range Nov 2025 – Nov 2026

64.6Anthropic60.0Google56.9OpenAI50.1Meta46.5Kimi