Models

Compare	Model	Context	Input /M	Output /M	Cached /M	Precision	Throughput	Latency	Capabilities	Uptime
	Gemini 3.8 Flash50% off Google	1M	$1.013 $2.026	$5.063 $10.126	—		—	—	+3	—
	Claude Fable 5.1 Anthropic	1M	$13.50	$67.50	—		—	—		—
	GPT Transcribe OpenAI	—	$0.0061 / min				—	—		—
	Gemini 3.5 Transcribe Google	—	$0.0067 / min				—	—		—
	Qwen 3.8 Flash Alibaba	1M	$0.203	$0.635	—		44tok/s	2.64s	+1	46.5%
	GLM 5.3 Flash50% off Zhipu AI	1M	$0.101 $0.203	$0.338 $0.675	—		30tok/s	2.50s	+1	91.2%
	DeepSeek V4 Flash Vision DeepSeek	1M	$0.297	$0.891	—		149tok/s	0.52s		100%
	GLM 5.3 Zhipu AI	1M	$1.89	$5.94	—		32tok/s	2.32s		96.4%
	Qwen 3.8 27B Alibaba	1M	$0.567	$4.05	—		50tok/s	1.55s		100%
	Gemini 3.7 Flash50% off Google	1M	$1.013 $2.026	$5.063 $10.126	$0.05063		91tok/s	2.98s	+4	96.6%
	Grok 4.6 xAI	500K	$2.70	$8.10	—		57tok/s	4.76s		97.0%
	Muse Glimmer 30B Meta	131K	$0.405	$1.62	—		57tok/s	5.01s		82.1%
	Muse Spark 1.2 Meta	1M	$1.688	$5.738	—		147tok/s	3.46s	+2	87.5%
	Qwen 3.8 Max Alibaba	1M	$2.2275	$6.684	$0.2781		28tok/s	2.96s		97.3%
	MiMo V2.5 Xiaomi	1M	$0.189	$0.378	—		24tok/s	3.27s		91.1%
	Qwen 3.7 Flash Alibaba	1M	$0.0498	$0.2164	$0.0081		108tok/s	2.51s	+2	87.9%
	Claude Opus 5 Fast Anthropic	1M	$13.50	$67.50	$1.35		33tok/s	1.44s	+2	95.4%
	Claude Opus 5 Anthropic	1M	$6.75	$33.75	$0.675		20tok/s	1.90s	+1	89.3%
	Gemini 3.6 Flash Google	1M	$1.013	$5.063	$0.10125		29tok/s	2.50s	+4	98.4%
	Gemini 3.5 Flash Lite Google	1M	$0.405	$3.375	$0.0405		57tok/s	1.07s	+4	96.2%
	Muse Spark 1.1 Meta	1M	$1.542	$5.243	—		178tok/s	2.02s	+2	94.1%
	Kimi K3 Moonshot	1M	$4.05	$20.25	$0.405		28tok/s	2.61s		87.5%
	GPT-5.6 Sol Pro50% off OpenAI	1M	$2.70 $5.40	$13.50 $27.00	$0.3375		52tok/s	3.82s	+1	95.5%
	GPT-5.6 Sol50% off OpenAI	1M	$2.70 $5.40	$13.50 $27.00	$0.3375		28tok/s	1.57s	+1	97.5%
	GPT-5.6 Terra Pro OpenAI	1M	$2.70	$16.20	$0.27		76tok/s	2.95s	+1	86.6%
	GPT-5.6 Terra OpenAI	1M	$2.70	$16.20	$0.27		37tok/s	1.22s	+1	95.8%
	GPT-5.6 Luna Pro OpenAI	1M	$0.27	$1.62	$0.027		77tok/s	3.00s	+2	92.3%
	GPT-5.6 Luna OpenAI	1M	$0.27	$1.62	$0.027		39tok/s	1.17s	+2	94.3%
	Grok 4.5 xAI	500K	$2.73	$8.189	—		48tok/s	2.48s		95.8%
	Hunyuan 3 Tencent	262K	$0.108	$0.446	—		71tok/s	3.76s		83.9%
	Claude Sonnet 5 Anthropic	1M	$2.70	$13.50	$0.27		18tok/s	4.25s	+1	96.8%
	GLM 5.2 Zhipu AI	1M	$1.101	$3.76	$0.29835	fp4 · fp8	78tok/s	2.81s		97.5%
	Kimi K2.7 Code Moonshot	262K	$1.009	$4.774	—		47tok/s	1.79s		99.1%
	Claude Fable 5 Anthropic	1M	$13.50	$67.50	—		16tok/s	3.59s		96.9%
	Gemini 3.5 Live Translate Google	—	$0.0635 / min				—	—		—
	Nemotron 3 Ultra 550B NVIDIA	1M	$0.675	$3.375	$0.135	fp4 · fp8	52tok/s	1.50s		98.0%
	Qwen 3.7 Plus Alibaba	1M	$0.4431	$1.773	$0.0864		54tok/s	16.79s		60.8%
	MiniMax M3 MiniMax	1M	$0.3985	$1.594	$0.0756	fp4 · fp8	34tok/s	2.11s	+1	97.3%
	Step 3.7 Flash StepFun	256K	$0.27	$1.553	$0.054	fp8	50tok/s	3.00s	+2	97.8%
	Qwen 3.7 Max Alibaba	1M	$2.304	$6.909	$0.16875		56tok/s	15.80s		79.4%
	Grok Build xAI	256K	$1.389	$2.777	$0.27		106tok/s	2.66s	+1	96.2%
	Gemini 3.5 Flash Google	1M	$2.025	$12.15	$0.2025		123tok/s	2.26s	+4	97.1%
	GPT Realtime Translate OpenAI	—	$0.0459 / min				—	—		—
	Grok 4.3 xAI	1M	$1.92	$3.838	$0.27		86tok/s	2.21s		96.1%
	DeepSeek V4 Pro DeepSeek	1M	$0.6901	$1.38	$0.133		32tok/s	3.54s		98.1%
	DeepSeek V4 Flash DeepSeek	1M	$0.1389	$0.2778	$0.0243		30tok/s	1.47s		99.9%
	GPT Image 2 OpenAI	—	$0.072 / image				—	—		98.7%
	Kimi K2.6 Moonshot	262K	$0.7856	$3.667	—		30tok/s	1.28s		97.2%
	Claude Opus 4.7 Anthropic	1M	$6.75	$33.75	$0.675		24tok/s	1.77s	+1	94.7%
	Seedance 2 Pro ByteDance	—	$0.4096 / sec				—	—		—
	Seedance 2 Fast ByteDance	—	$0.3266 / sec				—	—		—
	MiniMax Music Pro MiniMax	—	$0.2 / song				—	—		—
	GLM 5.1 Zhipu AI	203K	$1.89	$5.94	$0.27675	fp4 · fp8	29tok/s	11.59s		99.3%
	Qwen 3.6 Plus Alibaba	131K	$0.655	$3.93	—		51tok/s	8.58s		82.1%
	Gemma 4 31B Google	128K	$0.0763	$0.218	$0.135	fp4 · fp8	10tok/s	5.03s		99.9%
	Grok 4.20 Multi-Agent xAI	2M	$1.767	$3.534	—		308tok/s	7.50s		97.1%
	Grok 4.20 xAI	2M	$1.763	$3.528	—		39tok/s	1.09s		96.2%
	Gemini 3.1 Flash Live Google	—	$0.0389 / min				—	—		—
	MiniMax M2.7 MiniMax	205K	$0.405	$1.62	—		54tok/s	5.28s		91.6%
	Nano Banana 3 Flash Google	—	$0.061 / image				—	—		—
	Gemini 3.1 Pro Google	1M	$2.70	$16.20	$0.27		96tok/s	4.84s	+3	94.6%
	Recraft V4 Vector Recraft	—	$0.108 / image				—	—		—
	Claude Sonnet 4.6 Anthropic	1M	$4.05	$20.25	$0.405		31tok/s	1.94s	+1	96.5%
	Recraft V4 Recraft	—	$0.054 / image				—	—		—
	Recraft V4 Pro Recraft	—	$0.3375 / image				—	—		—
	Recraft V4 Vector Pro Recraft	—	$0.405 / image				—	—		—
	MiniMax M2.5 MiniMax	197K	$0.3826	$1.346	—		69tok/s	3.90s		99.8%
	Kling 3 Pro Kuaishou	—	$0.1512 / sec				—	—		—
	Kling 3 Pro (Audio) Kuaishou	—	$0.2268 / sec				—	—		—
	Kimi K2.5 Moonshot	262K	$0.6075	$3.038	—		35tok/s	5.54s		92.6%
	GLM 4.7 Flash Zhipu AI	203K	$0.081	$0.54	$0.0135	fp8	61tok/s	4.55s		95.7%
	Gemini 3 Flash Google	1M	$0.675	$4.05	$0.0675		40tok/s	1.42s	+4	98.7%
	Gemini 3 Flash (Audio) Google	—	$0.0026 / min				—	—		—
	Gemini 2.5 Flash Native Audio Google	—	$0.0389 / min				—	—		—
	DeepSeek V3 DeepSeek	160K	$0.351	$0.513	—	fp4	12tok/s	1.20s		83.1%
	FLUX.2 Pro Black Forest Labs	—	$0.0405 / image				—	—		—
	MiniMax Music MiniMax	—	$0.2 / song				—	—		—
	Claude Haiku 4.5 Anthropic	200K	$1.35	$6.75	$0.135		39tok/s	1.44s	+2	95.6%
	Kling 2.5 Pro Kuaishou	—	$0.0945 / sec				—	—		—
	EmbeddingGemma 300M Google	2K	$0.0027	$0.00	—		1182tok/s	0.34s		100%
	Nano Banana Google	—	$0.053 / image				—	—		—
	ElevenLabs Music ElevenLabs	—	$0.135 / sec				—	—		—
	GPT-OSS 120B OpenAI	128K	$0.0494	$0.2406	—		34tok/s	3.58s		92.4%
	Veo 3 Fast Google	—	$0.135 / sec				—	—		—
	GLM 4.5 Air Zhipu AI	131K	$0.1945	$1.272	$0.03375	fp8	37tok/s	3.10s		96.9%
	Qwen 3 Coder Alibaba	131K	$0.334	$1.519	$0.135	fp4 · fp8	48tok/s	1.18s		99.1%
	MiniMax Voice Design MiniMax	—	$2.00 / call				—	—		—
	Hailuo 02 (1080p) MiniMax	—	$0.78 / video				—	—		—
	Hailuo 02 (768p) MiniMax	—	$0.42 / video				—	—		—
	Hailuo 02 (512p) MiniMax	—	$0.14 / video				—	—		—
	ElevenLabs v3 ElevenLabs	—	$0.405 / 1K char				31ch/s	4.06s		—
	Qwen3 Embedding 8B Alibaba	33K	$0.0135	$0.00	—		197tok/s	2.04s		99.7%
	Qwen3 Embedding 0.6B Alibaba	33K	$0.0135	$0.00	—		1162tok/s	0.34s		100%
	Qwen3 Embedding 4B Alibaba	33K	$0.027	$0.00	—		1069tok/s	0.38s		100%
	Qwen3 Reranker 8B Alibaba	41K	$0.135	$0.00	—		—	—		—
	FLUX.1 Kontext Pro Black Forest Labs	—	$0.054 / image				—	—		—
	Veo 3 Google	—	$0.54 / sec				—	—		—
	Qwen 3 32B Alibaba	33K	$0.108	$0.378	—		46tok/s	8.95s		91.4%
	Llama 4 Maverick Meta	1M	$0.27	$1.08	—	fp8	33tok/s	1.13s		94.0%
	MiniMax Voice Clone MiniMax	—	$2.00 / call				—	—		—
	MiniMax Speech HD MiniMax	—	$0.07 / 1K char				39ch/s	3.25s		—
	MiniMax Speech Turbo MiniMax	—	$0.04 / 1K char				41ch/s	3.06s		100%
	Ideogram V3 Ideogram	—	$0.081 / image				—	—		—
	Gemini 2.5 Flash Google	1M	$0.405	$3.375	$0.0405		42tok/s	1.08s	+4	96.4%
	Sonar Pro Perplexity	200K	$4.05	$20.25	—		25tok/s	1.75s		95.3%
	Sonar Perplexity	127K	$1.35	$1.35	—		23tok/s	1.91s		97.2%
	MiniMax Image 01 MiniMax	—	$0.005 / image				—	—		14.9%
	DeepSeek R1 DeepSeek	64K	$0.7425	$2.957	—		66tok/s	1.88s		99.1%
	Llama 3.3 70B Meta	128K	$0.135	$0.432	—		15tok/s	0.95s		86.6%
	FLUX 1.1 Pro Ultra Black Forest Labs	—	$0.081 / image				—	—		—
	Recraft V3 Recraft	—	$0.054 / image				—	—		—
	ElevenLabs Flash v2.5 ElevenLabs	—	$0.2025 / 1K char				112ch/s	1.12s		—
	Whisper Large v3 Turbo OpenAI	—	$0.0009 / min				8.4×rt	1.20s		100%
	Hermes 3 405B Nous Research	131K	$1.35	$1.35	—	fp8	17tok/s	0.94s		100%
	Hermes 3 70B Nous Research	131K	$0.945	$0.945	—	fp8	25tok/s	0.85s		96.9%
	ElevenLabs Turbo v2.5 ElevenLabs	—	$0.2025 / 1K char				110ch/s	1.14s		—
	ElevenLabs Sound Effects ElevenLabs	—	$0.027 / call				—	—		—
	GPT-4o mini Transcribe OpenAI	—	$0.004 / min				4.9×rt	2.03s		100%
	Multilingual E5 Large Microsoft	1K	$0.0135	$0.00	—		430tok/s	1.19s		100%
	BGE-M3 BAAI	8K	$0.0135	$0.00	—		1289tok/s	0.41s		97.7%
	BGE Large EN v1.5 BAAI	1K	$0.0135	$0.00	—		1331tok/s	0.30s		100%
	BGE Base EN v1.5 BAAI	1K	$0.00675	$0.00	—		1391tok/s	0.29s		100%
	ElevenLabs Multilingual v2 ElevenLabs	—	$0.405 / 1K char				56ch/s	2.26s		—
	GTE Base Alibaba	1K	$0.00675	$0.00	—		442tok/s	0.91s		100%
	all-MiniLM-L12-v2 Sentence Transformers	1K	$0.00675	$0.00	—		11tok/s	11.58s		100%
	all-MiniLM-L6-v2 Sentence Transformers	1K	$0.00675	$0.00	—		285tok/s	0.90s		100%

Token models priced per 1M tokens. The cached column is what a repeated prompt prefix costs on that model — a rate of its own, read from the same source as the input rate beside it, not a fixed fraction of it. It runs from a tenth of the input rate to the full input rate depending on the model, and a dash means nobody publishes one, which is a different statement from “free”. Image, video, and audio models priced per unit (image, second, video, minute, song, call, or 1K characters). Throughput and latency are measured by Kyma every six hours with an identical input sent to every model through this API — so the numbers compare. For chat and embedding models throughput is tokens per second; for transcription models it is the realtime factor (×rt: seconds of audio transcribed per second of wall time); for text-to-speech it is characters synthesised per second (ch/s) — each off one fixed input. Image, video, music and realtime models are not probed; their reliability comes from real jobs, so those two columns stay empty by design. Uptime is every observation over 30 days, probes and real requests together. Precision names the numeric formats a model may be served at that are BELOW the one its creator released it in, and it is filled on the 14 models where that is true. A low format is not a downgrade when it is how the model shipped, so this is a comparison against the creator’s own model card rather than a list of small numbers. An empty cell means there is nothing to flag: either no route reports going under the release, or nobody could source what the release was. Those two are different answers and the model’s own page prints which one it is.