Offerings

Every active offering: a model as served by one host, with its current canonical price. Sort by clicking a column header; filter with the row below. Prices are USD per 1M tokens unless the unit toggle is set to per 1M characters (an approximation using a measured, blended tokens-per-character ratio).

Showing tier-tagged and vendor-direct models. Show all offerings, including the long tail

VendorHostTierOpen weightsShow free tierSearchUnit

Model	Vendor	Host	Variant	Input	Output	Blended ↑?	Cache read	Cache write	Batch in	Batch out	Context	Tokenizer	Last change	Confidence?	Source
Qwen3 235B A22B Instruct 2507	alibaba	nebius	–	0.200.0603	0.600.18	0.300.0905	–n/a	–n/a	–n/a	–n/a	131072	qwen2	–	aggregator	litellm
Codestral (latest)	mistral	google-vertex	–	0.200.0711	0.600.21	0.300.11	–n/a	–n/a	–n/a	–n/a	256000	mistral	–	aggregator	litellm
Llama 4 Maverick 17B 128E Instruct FP8	meta	openrouter	–	0.190.0564	0.650.20	0.300.0914	–n/a	–n/a	–n/a	–n/a	524288	llama3	–	api	openrouter
Llama 4 Maverick 17B 128E Instruct FP8	meta	digitalocean	unknown	0.190.0564	0.650.20	0.300.0914	–n/a	–n/a	–n/a	–n/a	524288	llama3	–	api	openrouter
Llama 4 Maverick 17B 128E Instruct FP8	meta	digitalocean	–	0.190.0564	0.650.20	0.300.0914	–n/a	–n/a	–n/a	–n/a	524288	llama3	–	api	openrouter
Gemini 2.5 Flash-Lite	google	google	unknown	0.18n/a	0.72n/a	0.32n/a	0.0180n/a	0.15n/a	–n/a	–n/a	1048576	gemini	–	api	openrouter
DeepSeek: DeepSeek V3	deepseek	deepseek	–	0.280.0918	0.420.14	0.320.10	0.02800.0092	–n/a	–n/a	–n/a	163840	deepseek	–	aggregator	litellm
DeepSeek-R1	deepseek	deepseek	–	0.280.0918	0.420.14	0.320.10	0.02800.0092	–n/a	–n/a	–n/a	163839	deepseek	–	aggregator	litellm
Gemini 2.5 Flash-Lite	google	google	–	0.18n/a	0.72n/a	0.32n/a	0.0180n/a	0.15n/a	–n/a	–n/a	1048576	gemini	–	api	openrouter
GPT-5 Mini	openai	openai	unknown	0.130.0352	1.000.28	0.340.0968	0.01250.0035	–n/a	–n/a	–n/a	400000	o200k	–	api	openrouter
GPT-5 Mini	openai	openai	–	0.130.0352	1.000.28	0.340.0968	0.01250.0035	–n/a	–n/a	–n/a	400000	o200k	–	api	openrouter
Llama 4 Scout Instruct	meta	azure	–	0.200.0602	0.780.23	0.350.10	–n/a	–n/a	–n/a	–n/a	131072	llama3	–	aggregator	litellm
Llama 4 Scout Instruct	meta	azure	–	0.200.0602	0.780.23	0.350.10	–n/a	–n/a	–n/a	–n/a	131072	llama3	–	aggregator	modelsdev
Llama 4 Maverick 17B 128E Instruct FP8	meta	deepinfra	–	0.200.0602	0.800.24	0.350.11	–n/a	–n/a	–n/a	–n/a	524288	llama3	–	aggregator	litellm
Qwen3 235B A22B Instruct 2507	alibaba	novita	–	0.200.0603	0.800.24	0.350.11	–n/a	–n/a	–n/a	–n/a	131072	qwen2	–	aggregator	litellm
Llama 4 Scout Instruct	meta	google-vertex	unknown	0.250.0752	0.700.21	0.360.11	–n/a	–n/a	–n/a	–n/a	131072	llama3	–	api	openrouter
Llama 4 Scout Instruct	meta	google-vertex	–	0.250.0752	0.700.21	0.360.11	–n/a	–n/a	–n/a	–n/a	131072	llama3	–	aggregator	litellm
Llama 4 Scout Instruct	meta	google-vertex	–	0.250.0752	0.700.21	0.360.11	–n/a	–n/a	0.130.0376	0.350.11	131072	llama3	–	aggregator	litellm
Llama 4 Scout Instruct	meta	google-vertex	–	0.250.0752	0.700.21	0.360.11	–n/a	–n/a	–n/a	–n/a	131072	llama3	–	api	openrouter
Qwen2.5-72B-Instruct	alibaba	deepinfra	–	0.360.11	0.400.12	0.370.11	–n/a	–n/a	–n/a	–n/a	131072	qwen2	–	aggregator	litellm
Qwen3 235B A22B Instruct 2507	alibaba	aws-bedrock	–	0.220.0663	0.880.27	0.390.12	–n/a	–n/a	0.110.0332	0.440.13	131072	qwen2	–	aggregator	litellm
Qwen3 235B A22B Instruct 2507	alibaba	fireworks	–	0.220.0663	0.880.27	0.390.12	–n/a	–n/a	–n/a	–n/a	131072	qwen2	–	aggregator	litellm
Qwen3 235B A22B Instruct 2507	alibaba	fireworks	–	0.220.0663	0.880.27	0.390.12	–n/a	–n/a	–n/a	–n/a	131072	qwen2	–	aggregator	litellm
Qwen3 235B A22B Instruct 2507	alibaba	aws-bedrock	–	0.220.0663	0.880.27	0.390.12	–n/a	–n/a	–n/a	–n/a	131072	qwen2	–	aggregator	litellm
GPT-6 Luna (Global)	openai	openai	unknown	0.200.0563	1.000.28	0.400.11	0.02000.0056	0.250.0704	–n/a	–n/a	1050000	o200k	–	api	openrouter
DeepSeek-R1	deepseek	hyperbolic	–	0.400.13	0.400.13	0.400.13	–n/a	–n/a	–n/a	–n/a	163839	deepseek	–	aggregator	litellm
GPT-6 Luna (Global)	openai	openai	fast	0.200.0563	1.000.28	0.400.11	0.02000.0056	0.250.0704	–n/a	–n/a	1050000	o200k	–	official	official-openai
GPT-6 Luna (Global)	openai	openai	–	0.200.0563	1.000.28	0.400.11	0.02000.0056	0.250.0704	–n/a	–n/a	1050000	o200k	–	api	openrouter
Llama 4 Maverick 17B 128E Instruct FP8	meta	novita	fp8	0.270.0813	0.850.26	0.420.12	–n/a	–n/a	–n/a	–n/a	524288	llama3	–	api	openrouter
Llama 4 Maverick 17B 128E Instruct FP8	meta	novita	–	0.270.0813	0.850.26	0.420.12	–n/a	–n/a	–n/a	–n/a	524288	llama3	–	aggregator	litellm
Gemini 2.5 Flash	google	google	unknown	0.15n/a	1.25n/a	0.43n/a	0.0150n/a	0.0417n/a	–n/a	–n/a	1048576	gemini	–	api	openrouter
DeepSeek: DeepSeek V3	deepseek	deepinfra	–	0.250.0820	0.950.31	0.430.14	0.220.0708	–n/a	–n/a	–n/a	163840	deepseek	–	aggregator	litellm
Gemini 2.5 Flash	google	google	–	0.15n/a	1.25n/a	0.43n/a	0.0150n/a	0.0417n/a	–n/a	–n/a	1048576	gemini	–	api	openrouter
Llama 4 Maverick 17B 128E Instruct FP8	meta	azure	–	0.250.0752	1.000.30	0.440.13	–n/a	–n/a	–n/a	–n/a	524288	llama3	–	aggregator	litellm
Llama 4 Maverick 17B 128E Instruct FP8	meta	azure	–	0.250.0752	1.000.30	0.440.13	–n/a	–n/a	–n/a	–n/a	524288	llama3	–	aggregator	modelsdev
Codestral (latest)	mistral	azure	–	0.300.11	0.900.32	0.450.16	–n/a	–n/a	–n/a	–n/a	256000	mistral	–	aggregator	litellm
Codestral (latest)	mistral	azure	–	0.300.11	0.900.32	0.450.16	–n/a	–n/a	–n/a	–n/a	256000	mistral	–	aggregator	modelsdev
DeepSeek: DeepSeek V3	deepseek	openrouter	–	0.260.0844	1.030.34	0.450.15	–n/a	–n/a	–n/a	–n/a	163840	deepseek	–	api	openrouter
DeepSeek: DeepSeek V3	deepseek	streamlake	unknown	0.260.0844	1.030.34	0.450.15	–n/a	–n/a	–n/a	–n/a	163840	deepseek	–	api	openrouter
DeepSeek: DeepSeek V3	deepseek	streamlake	–	0.260.0844	1.030.34	0.450.15	–n/a	–n/a	–n/a	–n/a	163840	deepseek	–	api	openrouter
DeepSeek: DeepSeek V3	deepseek	deepinfra	fp4	0.320.10	0.890.29	0.460.15	–n/a	–n/a	–n/a	–n/a	163840	deepseek	–	api	openrouter
Qwen3 Coder 480B A35B Instruct	alibaba	deepinfra	–	0.300.0905	1.000.30	0.480.14	0.100.0302	–n/a	–n/a	–n/a	262144	qwen2	–	aggregator	litellm
Grok 3 mini	xai	azure	–	0.25n/a	1.27n/a	0.51n/a	–n/a	–n/a	–n/a	–n/a	131072	–	–	aggregator	litellm
Llama 4 Maverick 17B 128E Instruct FP8	meta	parasail	fp8	0.350.11	1.000.30	0.510.15	0.170.0512	–n/a	–n/a	–n/a	524288	llama3	–	api	openrouter
Llama 4 Maverick 17B 128E Instruct FP8	meta	google-vertex	unknown	0.350.11	1.150.35	0.550.17	–n/a	–n/a	–n/a	–n/a	524288	llama3	–	api	openrouter
Llama 4 Maverick 17B 128E Instruct FP8	meta	google-vertex	–	0.350.11	1.150.35	0.550.17	–n/a	–n/a	0.180.0527	0.580.17	524288	llama3	–	aggregator	litellm
Llama 4 Maverick 17B 128E Instruct FP8	meta	google-vertex	–	0.350.11	1.150.35	0.550.17	–n/a	–n/a	–n/a	–n/a	524288	llama3	–	aggregator	litellm
Llama 4 Maverick 17B 128E Instruct FP8	meta	google-vertex	–	0.350.11	1.150.35	0.550.17	–n/a	–n/a	–n/a	–n/a	524288	llama3	–	api	openrouter
Llama 3.3 70B Instruct	meta	sambanova	unknown	0.450.14	0.900.27	0.560.17	–n/a	–n/a	–n/a	–n/a	131072	llama3	–	api	openrouter
Llama 3.3 70B Instruct	meta	sambanova	–	0.450.14	0.900.27	0.560.17	–n/a	–n/a	–n/a	–n/a	131072	llama3	–	api	openrouter

Confidence: official = vendor pricing page · api = OpenRouter per-host · aggregator = community list, may lag · unreviewed = automatic extraction awaiting review. When sources disagree the most authoritative wins. Details: /methodology#confidence