Leaderboard | Coding

Models

Choose model family

Claude by Anthropic

Mistral by Mistral AI

More filters

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.

All Single turn Multiple turns

Open license models

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

All selected Open license Proprietary license

876

DeepSeek R1T2 Chimera

875

Mistral Medium 3

874

Solar Mini 250422

871

GLM 4.7 Flash

871

Mixtral 8x22B

871

Devstral Small 2505

869

AFM 4.5B

869

Claude Sonnet 3

868

Krutrim Spectre V2

868

GLM 4 32B

867

Gemma 3 12B

863

Moonshot V1 8k

861

Qwen 2.5 14B Instruct

856

DeepSeek-R1 0528 Qwen3 8B

854

Ministral 3B 2512

Last updated about 1 month ago

Rank	Overall	Name	VIBE Score	Confidence Interval	Votes	Downvote %	Abort %	Speed	Latency	Context	Cost (Input)	Cost (Output)
241	210	DeepSeek R1T2 Chimera	876	±10	2.1K	5.9%	3.0%	28 tps	1.8s	164K	$0.13	$0.45
242	210	Mistral Medium 3	875	±23	485	6.7%	2.4%	47 tps	0.8s	33K	$0.40	$2.00
243	210	Solar Mini 250422	874	±17	1.3K	5.9%	1.8%	90 tps	1.7s	33K	$0.15	$0.15
244	210	GLM 4.7 Flash	871	±28	610	4.7%	5.8%	61 tps	2.8s	128K	$0.07	$0.39
245	210	Mixtral 8x22B	871	±22	1.3K	5.0%	1.2%	140 tps	0.6s	64K	$2.00	$6.00
246	293	Devstral Small 2505	871	±15	1.7K	6.2%	<0.1%	141 tps	1.3s	33K	$0.03	$0.09
247	293	AFM 4.5B	869	±7	4.4K	8.9%	<0.1%	81 tps	0.3s	66K	$0.05	$0.20
248	293	Claude Sonnet 3	869	±17	900	1.6%	<0.1%	35 tps	1.0s	200K	$3.00	$15.00
249	210	Krutrim Spectre V2	868	±16	1.3K	3.6%	<0.1%	33 tps	3.1s	4K	$0.19	$0.19
250	210	GLM 4 32B	868	±12	2.9K	4.9%	2.6%	40 tps	1.6s	33K	$0.14	$0.14
251	210	Gemma 3 12B	867	±11	2.5K	4.9%	4.2%	73 tps	0.8s	131K	$0.05	$0.12
252	210	Moonshot V1 8k	863	±13	915	5.2%	1.0%	55 tps	1.5s	8K	$0.20	$2.00
253	210	Qwen 2.5 14B Instruct	861	±16	2.4K	5.7%	2.4%	40 tps	1.6s	1M	$0.40	$1.61
254	312	DeepSeek-R1 0528 Qwen3 8B	856	±8	4.9K	6.5%	<0.1%	45 tps	2.4s	128K	$0.05	$0.09
255	210	Ministral 3B 2512	854	±57	515	8.0%	2.8%	339 tps	0.6s	131K	$0.10	$0.10
256	234	Jamba 1.5 Large	851	±9	2.9K	4.0%	1.7%	48 tps	0.9s	256K	$1.50	$6.00
257	312	Wikipedia	846	±7	9.8K	4.9%	<0.1%	47 tps	2.1s	32K	$0	$0
258	324	MAI-DS-R1	842	±7	3.5K	11.7%	<0.1%	73 tps	3.2s	64K	$0.10	$0.40
259	234	GPT-3.5 Turbo 16k	838	±10	2.7K	3.6%	<0.1%	22 tps	0.6s	16K	$3.00	$4.00
260	324	Cogito V2 671B	838	±17	1.6K	5.9%	<0.1%	41 tps	0.6s	164K	$1.25	$1.25
261	234	ERNIE 4.5 21B A3B Thinking	838	±23	1.1K	6.9%	1.8%	87 tps	1.5s	120K	$0.07	$0.28
262	240	GLM 4.5 Flash	834	±37	520	8.8%	12.2%	15 tps	2.2s	131K	$0	$0
263	324	NVIDIA Llama 3.1 Nemotron Ultra 253B v1	832	±16	2.2K	4.1%	<0.1%	40 tps	0.8s	128K	$0.30	$0.90
264	240	LFM2 2.6B	826	±26	810	10.0%	6.7%	184 tps	0.4s	33K	$0.01	$0.02
265	240	Krutrim 2	825	±10	2.3K	2.3%	12.5%	33 tps	2.1s	128K	$1.00	$1.00
266	240	Moonshot V1 32k	820	±17	950	3.1%	1.4%	53 tps	1.4s	33K	$1.00	$3.00
267	240	LFM2 8B A1B	818	±18	825	11.3%	<0.1%	142 tps	0.3s	33K	$0.01	$0.02
268	337	GLM 4.1V 9B Thinking	813	±16	1.1K	4.2%	<0.1%	69 tps	1.3s	66K	$0.04	$0.14
269	337	Qwen 2 72B Instruct	805	±19	1K	3.3%	<0.1%	3 tps	N/A	33K	$0.90	$0.90
270	346	Magistral Medium (Thinking)	804	±10	2.2K	5.7%	<0.1%	67 tps	0.8s	41K	$2.00	$5.00
271	252	Magistral Small 2509	802	±18	1.8K	7.5%	2.7%	116 tps	0.6s	131K	$0.50	$1.50
272	252	WizardLM-2 8x22B	801	±12	1.9K	3.1%	11.6%	11 tps	2.5s	66K	$0.77	$0.77
273	252	Hermes 4 405B FP8	797	±21	815	8.4%	3.5%	31 tps	0.9s	131K	$0.52	$1.73
274	346	Magistral Medium 2507	795	±25	665	14.2%	<0.1%	86 tps	0.7s	41K	$2.00	$5.00
275	252	Mercury Coder	793	±27	510	3.8%	<0.1%	247 tps	2.2s	32K	$0.25	$1.00
276	252	GPT-3.5 Turbo Instruct	787	±9	2K	2.7%	<0.1%	46 tps	1.2s	4K	$1.50	$2.00
277	252	Mistral Large	785	±16	1.1K	5.8%	1.5%	54 tps	0.7s	33K	$2.00	$6.00
278	252	Hermes 4 70B	781	±29	460	8.9%	1.1%	67 tps	0.6s	131K	$0.12	$0.39
279	262	Baichuan-M2-32B	770	±30	740	10.8%	<0.1%	32 tps	3.3s	131K	$0.07	$0.07
280	354	OLMo 3 7B Think	763	±21	770	7.8%	4.2%	77 tps	0.4s	66K	$0.12	$0.20

7of8

View All (305 models)