Text Leaderboard

Overall rankings for text generation models, powered by community votes and the Elo rating system.

Scroll horizontally to see all columns

Text

#	Model(14/14)	Elo	Votes	$/1M In / Out	Value
1	Gemini 3 Flash Preview	1268	14	$0.50 / $3.00	725
2	Functionary Swahili Large	1262	26	$0.20 / $1.50	1,485
3	GLM 5	1254	21	$0.72 / $2.30	830
4	Claude Sonnet 4.5	1244	14	$3.00 / $15.00	138
5	GPT-oss-120B	1241	13	$0.04 / $0.19	10,838
6	Claude Sonnet 4.6	1233	15	$3.00 / $15.00	137
7	Grok 4.1 Fast	1208	24	$0.20 / $0.50	3,451
8	Rnj 1 Instruct	1193	8	$0.15 / $0.15	7,953
9	Functionary Swahili Mini	1191	20	$0.10 / $0.90	2,382
10	Claude Haiku 4.5	1182	19	$1.00 / $5.00	394
11	Gemini 2.5 Flash Lite	1178	15	$0.10 / $0.40	4,712
12	GPT-5.2	1155	18	$1.75 / $14.00	147
13	Trinity Large Preview	1140	15	$0.00 / $0.00	-
14	GPT-5 Nano	1120	21	$0.05 / $0.40	4,978

Uniform sampling, excluding ties

Each cell shows how often the row model beats the column model

Loading head-to-head data…

Approximate 95% confidence intervals — narrower bars = more certain ranking

Total number of head-to-head challenges between each model pair

Loading challenge count data…