Model rankings

Compare model usage and performance from all Kyma API traffic, including Kyma's own accounts.

Live data4.4B tokens across 100 models from 3,694 developers

100 models = has at least one request in the window; 106 models = full live catalog.

#	Model	Tokens	Speed	Uptime
1	deepseek-v4-flashby DeepSeek	1.2B	28tok/s	100%
2	qwen-3.6-plusby Alibaba	550.7M	50tok/s	100%
3	gemma-4-31bby Google	395.8M	15tok/s	100%
4	deepseek-v4.1-flashby DeepSeek	342.0M	46tok/s	100%
5	deepseek-v3by DeepSeek	183.5M	10tok/s	97%
6	gpt-oss-120bby OpenAI	131.7M	16tok/s	100%
7	minimax-m2.7by MiniMax	119.8M	13tok/s	100%
8	minimax-m2.5by MiniMax	102.9M	18tok/s	97%
9	gemini-3-flashby Google	97.7M	73tok/s	99%
10	qwen-3-coderby Alibaba	95.8M	6tok/s	100%
11	glm-5.3-flashby Zhipu AI	89.9M	38tok/s	100%
12	gemini-3.7-flashby Google	83.4M	94tok/s	98%
13	llama-3.3-70bby Meta	74.9M	6tok/s	87%
14	qwen3.7-flashby Alibaba	62.2M	73tok/s	97%
15	gpt-5.6-lunaby OpenAI	57.0M	75tok/s	99%
16	glm-4.7-flashby Zhipu AI	53.5M	9tok/s	100%
17	kimi-k3by Moonshot	45.8M	55tok/s	100%
18	qwen-3.8-maxby Alibaba	45.2M	17tok/s
19	gemini-3.5-flashby Google	45.0M	108tok/s	96%
20	qwen3.8-flashby Alibaba	42.1M	13tok/s	88%
21	deepseek-v4-flash-vision-expby DeepSeek	41.4M	60tok/s	97%
22	qwen-3-32bby Alibaba	38.4M	13tok/s	94%
23	glm-5.3by Zhipu AI	29.5M	54tok/s	100%
24	minimax-m3by MiniMax	27.2M	10tok/s	100%
25	deepseek-r1by DeepSeek	23.6M	21tok/s	100%
26	gemini-3.5-flash-liteby Google	16.5M	77tok/s	100%
27	glm-4.5-airby Zhipu AI	15.5M	21tok/s	100%
28	muse-spark-1.2by Meta	15.0M	108tok/s
29	glm-5.1by Zhipu AI	15.0M	27tok/s
30	gpt-5.6-solby OpenAI	13.9M	57tok/s	98%
31	kimi-k2.6by Moonshot	12.7M	12tok/s	100%
32	gpt-5.6-luna-proby OpenAI	12.6M	22tok/s	97%
33	glm-5.2by Zhipu AI	11.3M	73tok/s	100%
34	qwen3.8-27bby Alibaba	10.0M	17tok/s	98%
35	gemini-3.6-flashby Google	8.9M	73tok/s	87%
36	grok-4.20-multi-agentby xAI	6.2M	320tok/s
37	qwen-3.7-maxby Alibaba	5.8M	104tok/s
38	claude-fable-5by Anthropic	5.7M	9tok/s
39	grok-4.3by xAI	4.5M	73tok/s
40	qwen-3.7-plusby Alibaba	4.2M	21tok/s	100%
41	claude-sonnet-5by Anthropic	3.9M	74tok/s	100%
42	gemini-3.8-flashby Google	3.9M	57tok/s	99%
43	kimi-k2.7-codeby Moonshot	3.1M	25tok/s
44	gpt-5.6-terra-proby OpenAI	3.0M	38tok/s
45	gpt-5.6-sol-proby OpenAI	2.6M	32tok/s
46	claude-haiku-4-5by Anthropic	1.7M	158tok/s	100%
47	step-3.7-flashby StepFun	1.3M	11tok/s	100%
48	grok-buildby xAI	1.2M	85tok/s
49	claude-opus-5by Anthropic	1.2M	13tok/s
50	claude-sonnet-4-6by Anthropic	944.8K	39tok/s	100%
51	claude-fable-5.1by Anthropic	942.3K	11tok/s
52	sonarby Perplexity	899.0K	131tok/s	100%
53	gpt-5.6-terraby OpenAI	840.4K	13tok/s
54	llama-4-maverickby Meta	790.5K	7tok/s	100%
55	whisper-v3-turboby OpenAI	677.3K		99%
56	nemotron-3-ultra-550bby NVIDIA	564.3K	39tok/s
57	grok-4.5by xAI	485.8K	37tok/s	100%
58	grok-4.6by xAI	344.7K	35tok/s
59	kling-3-proby Kuaishou	330.8K
60	muse-spark-1.1by Meta	316.1K	111tok/s
61	mimo-v2.5by Xiaomi	265.1K	11tok/s	100%
62	gemini-3.1-proby Google	213.3K	64tok/s
63	grok-4.20by xAI	211.6K	10tok/s
64	grok-4.7by xAI	187.5K	24tok/s	100%
65	claude-opus-4-7by Anthropic	171.3K	20tok/s
66	qwen3-embedding-8bby Alibaba	150.7K		100%
67	gpt-6-astraby OpenAI	136.1K	8tok/s	4%
68	hy3by Tencent	127.8K	8tok/s	100%
69	ideogram-v3by Ideogram	111.8K
70	flux-kontext-proby Black Forest Labs	104.1K
71	hermes-3-70bby Nous Research	88.5K	10tok/s
72	bge-m3by BAAI	72.9K
73	muse-glimmer-30bby Meta	72.7K	28tok/s	100%
74	multilingual-e5-largeby Microsoft	65.2K
75	sonar-proby Perplexity	59.5K	9tok/s
76	hermes-3-405bby Nous Research	56.2K	10tok/s
77	jevby TypeSafe	55.5K		100%
78	embeddinggemma-300mby Google	53.0K
79	gte-baseby Alibaba	51.3K
80	bge-large-enby BAAI	50.7K
81	bge-base-enby BAAI	50.7K
82	qwen3-embedding-0.6bby Alibaba	50.6K
83	qwen3-embedding-4bby Alibaba	50.6K
84	all-minilm-l6by Sentence Transformers	33.2K
85	gpt-4o-mini-transcribe-2025-12-15by OpenAI	18.9K		100%
86	all-minilm-l12by Sentence Transformers	16.3K
87	muse-spark-1.3by Meta	15.2K	57tok/s
88	qwen3-reranker-8bby Alibaba	11.9K
89	eleven-flash-v2-5by ElevenLabs	112
90	recraft-v4-proby Recraft	56
91	eleven-turbo-v2-5by ElevenLabs	56
92	minimax-speech-turboby MiniMax	56
93	minimax-image-01by MiniMax	56		98%
94	recraft-v4-vectorby Recraft	56
95	recraft-v4by Recraft	56
96	flux-2-proby Black Forest Labs	56
97	eleven-multilingual-v2by ElevenLabs	56		100%
98	minimax-speech-hdby MiniMax	56
99	gpt-transcribeby OpenAI	0		100%
100	gemini-3.5-transcribeby Google	0		91%

Measured from real API traffic. Updated every 5 minutes. Read docs →