Leaderboard | Coding

Show inactive models

Hide models that are no longer actively available on Yupp.

Turns

Filter model performance by the number of turns in a conversation.

Open license models

Filter the leaderboard to only show models that have an open license.

1059

Llama 3 8B Turbo

1037

Llama 3 70B Turbo

971

Llama 4 Maverick

958

Llama 3.1 8B Turbo

942

NVIDIA Llama 3.3 Nemotron Super 49B v1

928

NVIDIA Llama 3.1 Nemotron 70B

896

Llama 3.1 405B Instruct Turbo

885

Llama 3.2 11B Instruct

864

Hermes 2 Pro Llama 3 8B

851

Llama 3.3 70B Instruct Turbo

835

DeepSeek-R1 Distill Llama 70B

Last updated about 1 month ago

Rank	Overall	Name	VIBE Score	Confidence Interval	Votes	Downvote %	Abort %	Speed	Latency	Context	Cost (Input)	Cost (Output)
1	151	Llama 3 8B Turbo	1059	±24	600	1.6%	<0.1%	97 tps	0.1s	8K	$0.12	$0.13
2	164	Llama 3 70B Turbo	1037	±6	4.3K	1.0%	<0.1%	31 tps	0.0s	8K	$0.73	$0.83
3	167	Llama 4 Maverick	971	±5	21K	5.0%	1.2%	88 tps	2.4s	1M	$0.23	$0.83
4	167	Llama 3.1 8B Turbo	958	±14	2.2K	2.0%	2.1%	650 tps	0.5s	128K	$0.13	$0.14
5	230	NVIDIA Llama 3.3 Nemotron Super 49B v1	942	±9	3.6K	2.3%	<0.1%	13 tps	N/A	131K	$0.07	$0.20
6	245	NVIDIA Llama 3.1 Nemotron 70B	928	±7	5.3K	2.0%	<0.1%	9 tps	0.1s	128K	$0.33	$0.39
7	264	Llama 3.1 405B Instruct Turbo	896	±11	2K	3.9%	<0.1%	26 tps	0.8s	131K	$3.50	$3.50
8	201	Llama 3.2 11B Instruct	885	±15	2.1K	4.1%	1.5%	152 tps	0.5s	8K	$0.16	$0.16
9	210	Hermes 2 Pro Llama 3 8B	864	±21	1.8K	2.5%	<0.1%	76 tps	1.0s	131K	$0.08	$0.09
10	234	Llama 3.3 70B Instruct Turbo	851	±19	1.2K	6.0%	2.0%	78 tps	1.0s	131K	$0.88	$0.88
11	240	DeepSeek-R1 Distill Llama 70B	835	±9	3.4K	5.2%	3.6%	27 tps	1.6s	32K	$0.73	$0.95