Text Model Leaderboard

Combine overall conclusions with third-party evidence to choose models for coding, reasoning, and professional work.


1	ANClaude Opus 5.5	Sep 22, 2026	100%27 items	$5	$4	$20	78.3
2	ANClaude Sonnet 5.5	Sep 28, 2026	100%27 items	$2.5	$2	$10	75.2
3	OPGPT-6 Astra	Sep 3, 2026	100%27 items	$1	$10	$50	74.7
4	GOGemini 4 Argon	Sep 30, 2026	100%27 items	$0.1	$2	$10	73.3
5	ANClaude Fable 5.1	Sep 1, 2026	97%26 items	$12.5	$10	$50	72.8
6	OPGPT-6.1 Sol	Oct 8, 2026	100%27 items	$0.1	$2	$10	71.5
7	ANClaude Opus 5	Jul 24, 2026	97%26 items	$6.25	$5	$25	67.4
8	OPGPT-6 Sol	Sep 22, 2026	94%25 items	$0.2	$2	$10	65.2
9	OPGPT-6.1 Terra	Oct 26, 2026	69%18 items	$0.05	$1	$5	64.9
10	ANClaude Fable 5	Jun 9, 2026	83%22 items	$12.5	$10	$50	64.8
11	OPGPT-5.6 Sol	Jul 9, 2026	92%25 items	$0.4	$4	$20	64.2
12	MEMuse Spark 1.3	Sep 2, 2026	87%24 items	$0.15	$1.25	$4.25	60.6
13	ANClaude Haiku 5.5	Oct 7, 2026	100%27 items	$0.125	$0.1	$0.5	60.5
14	DEDeepSeek V4.1 Flash 1017	Oct 17, 2026	81%21 items	$0.003	$0.15	$0.6	60.5
15	ZAGLM 5.3	Oct 5, 2026	100%27 items	$0.26	$1.4	$4.4	60.3
16	OPGPT-5.5	Apr 23, 2026	74%20 items	$0.5	$5	$30	59.8
17	XAGrok 4.7	Sep 21, 2026	95%26 items	$0.5	$2	$6	59.1
18	GOGemini 3.8 Flash	Sep 2, 2026	95%26 items	$0.075	$0.75	$3.75	59.0
19	OPGPT-5.6 Terra	Jul 9, 2026	91%24 items	$0.2	$2	$12	59.0
20	ANClaude Opus 4.8	May 28, 2026	83%22 items	$6.25	$5	$25	58.8
21	XAGrok 4.6	Aug 12, 2026	97%26 items	$0.5	$2	$6	58.3
22	ALQwen3.8 Max	Aug 3, 2026	86%23 items	$0.25	$2	$6	57.4
23	XIMiMo V2.6 Pro	Sep 21, 2026	78%21 items	$0.0036	$0.435	$0.87	57.2
24	STStep 5 Preview	Sep 20, 2026	69%19 items	¥0.35	¥7	¥20	57.0
25	DEDeepSeek V4.1 Flash	Sep 10, 2026	92%25 items	$0.003	$0.15	$0.6	56.1
26	MAKimi K3	Jul 16, 2026	97%26 items	$3	$3	$15	55.6
27	MIMiniMax M3.1 Flash	Oct 19, 2026	58%15 items	$0.015	$0.15	$0.6	55.1
28	ANClaude Sonnet 5	Jun 30, 2026	83%22 items	$2.5	$2	$10	54.7
29	XAGrok 4.5	Jul 8, 2026	76%20 items	$0.3	$2	$6	53.9
30	OPGPT-6 Luna	Sep 22, 2026	100%27 items	$0.01	$0.1	$0.5	53.8
31	DEDeepSeek V4 Pro 0813	Aug 13, 2026	89%23 items	$0.022	$0.66	$1.98	53.7
32	MEMuse Spark 1.2	Aug 5, 2026	70%18 items	$0.15	$1.25	$4.25	53.6
33	ZAGLM-5.3 Flash	Aug 26, 2026	87%24 items	$0.03	$0.15	$0.5	53.5
34	OPGPT-5.6 Luna	Jul 9, 2026	94%25 items	$0.02	$0.2	$1.2	52.9
35	GOGemini 3.6 Flash	Jul 21, 2026	74%20 items	$0.075	$0.75	$3.75	52.9
36	MEMuse Spark 1.1	Apr 8, 2026	60%16 items	$0.15	$1.25	$4.25	52.8
37	DEDeepSeek V4 Flash 0731	Jul 31, 2026	71%18 items	Unverified	$0.016	$0.64	52.8
38	ZAGLM-5.2	Jun 13, 2026	71%19 items	$0.26	$1.4	$4.4	51.4
39	ANClaude Sonnet 4.6	Feb 17, 2026	71%19 items	$3.75	$3	$15	50.1
40	ALQwen3.7 Max	May 21, 2026	66%17 items	$0.5	$2.5	$7.5	49.4
41	MAKimi K2.6	Apr 21, 2026	68%18 items	$0.16	$0.95	$4	48.9
42	MAKimi K2.7 Code	Jun 12, 2026	60%15 items	$0.19	$0.95	$4	47.8
43	XIMiMo-V2.5-Pro	Apr 22, 2026	57%15 items	$0.0036	$0.435	$0.87	47.2
44	XAGrok 4.3	Apr 17, 2026	57%15 items	$0.2	$1.25	$2.5	46.6
45	ZAGLM-5.1	Apr 7, 2026	63%17 items	$0.26	$1.4	$4.4	46.5
46	OPGPT-5.4 mini	Mar 17, 2026	76%20 items	$0.075	$0.75	$4.5	46.1
47	XIMiMo-V2.5	Apr 22, 2026	44%12 items	$0.0028	$0.14	$0.28	45.8
48	ALQwen3.7 Plus	Jun 2, 2026	50%14 items	$0.05	$0.4	$1.6	45.7
49	MIMiniMax-M3	Jun 1, 2026	77%21 items	$0.06	$0.3	$1.2	44.7
50	TMInkling Small	Jul 30, 2026	79%21 items	Unverified	$0.45	$1.2	44.3
51	MIMiniMax-M2.7	Mar 18, 2026	57%15 items	$0.06	$0.3	$1.2	44.2
52	GOGemini 3.5 Flash Lite	Jul 21, 2026	77%21 items	$0.03	$0.3	$2.5	43.6
53	TMInkling	Jul 15, 2026	86%23 items	$0.374	$1.87	$4.68	43.4
54	INLing-3.0-flash (free)	Jul 23, 2026	44%11 items	Unverified	$0.021	$0.063	42.8
55	ANClaude Haiku 4.5	Oct 15, 2025	73%19 items	$1.25	$1	$5	41.5

Not yet ranked

These models have partial scores and will receive a rank when evidence is complete.

View ranking requirements


MEMuse Wonder Full	66.2	Only one evaluation institution so far · Fewer than 3 coding evaluation types
MEMuse Wonder	64.9	Fewer than 3 coding evaluation types
XAGrok Voice 3.0	25.0	Only one evaluation institution so far · Fewer than 3 coding evaluation types · Fewer than 2 reasoning evaluation types · Fewer than 2 professional-work evaluation types · Fewer than 2 knowledge and language evaluation types
IGGranite Switch 4.1 8B	25.0	Only one evaluation institution so far · Fewer than 3 coding evaluation types · Fewer than 2 reasoning evaluation types · Fewer than 2 professional-work evaluation types · Fewer than 2 knowledge and language evaluation types
IGGranite Switch 4.1 30B	25.0	Only one evaluation institution so far · Fewer than 3 coding evaluation types · Fewer than 2 reasoning evaluation types · Fewer than 2 professional-work evaluation types · Fewer than 2 knowledge and language evaluation types
NVNVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16	25.0	Only one evaluation institution so far · Fewer than 3 coding evaluation types · Fewer than 2 reasoning evaluation types · Fewer than 2 professional-work evaluation types · Fewer than 2 knowledge and language evaluation types
NVNVIDIA-Nemotron-3.5-Super-VL-120B-A12B-NVFP4	25.0	Only one evaluation institution so far · Fewer than 3 coding evaluation types · Fewer than 2 reasoning evaluation types · Fewer than 2 professional-work evaluation types · Fewer than 2 knowledge and language evaluation types

Text Model Leaderboard

Not yet ranked

Frequently asked questions

Why are some catalog models not ranked?

Is the overall capability score an accuracy percentage?

Where can I see individual evaluation results?

What are the weights in the overall score?

How do raw evaluation results become an overall score?

What happens to the weights when a dimension has no results?

What evidence is required for a formal rank?