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

ModelVendorHostVariantInputOutputBlended ↑?Cache readCache writeBatch inBatch outContextTokenizerLast changeConfidence?Source
Llama 3.3 70B Instructmetaprimeintellectunknown1.040.312.250.681.340.40–n/a–n/a–n/a–n/a131072llama3–apiopenrouter
Llama 3.3 70B Instructmetaprimeintellect–1.040.312.250.681.340.40–n/a–n/a–n/a–n/a131072llama3–apiopenrouter
Nova Pro (EU)amazonaws-bedrock–0.80n/a3.20n/a1.40n/a0.20n/a–n/a–n/a–n/a300000––aggregatorlitellm
Nova Pro (EU)amazonamazon–0.80n/a3.20n/a1.40n/a–n/a–n/a–n/a–n/a300000––aggregatorlitellm
Nova Pro (EU)amazonaws-bedrock–0.80n/a3.20n/a1.40n/a0.20n/a–n/a–n/a–n/a300000––aggregatorlitellm
Nova Pro (EU)amazonaws-bedrock–0.84n/a3.36n/a1.47n/a0.21n/a–n/a–n/a–n/a300000––aggregatorlitellm
Kimi K2 Instructmoonshottogether–1.00n/a3.00n/a1.50n/a–n/a–n/a–n/a–n/a262144––aggregatorlitellm
Gemini 2.5 Flashgooglegoogle-vertexunknown0.54n/a4.50n/a1.53n/a0.0540n/a0.15n/a–n/a–n/a1048576gemini–apiopenrouter
Gemini 2.5 Flashgooglegoogleunknown0.54n/a4.50n/a1.53n/a0.0540n/a0.15n/a–n/a–n/a1048576gemini–apiopenrouter
Gemini 2.5 Flashgooglegoogle-vertex–0.54n/a4.50n/a1.53n/a0.0540n/a0.15n/a–n/a–n/a1048576gemini–apiopenrouter
Gemini 2.5 Flashgooglegoogle–0.54n/a4.50n/a1.53n/a0.0540n/a0.15n/a–n/a–n/a1048576gemini–apiopenrouter
Qwen3 Maxalibabaopenrouter–0.780.243.901.181.560.470.160.04700.980.29–n/a–n/a262144qwen2–apiopenrouter
Qwen3 Maxalibabaalibabaunknown0.780.243.901.181.560.470.160.04700.980.29–n/a–n/a262144qwen2–apiopenrouter
Qwen3 Maxalibabaalibaba–0.780.243.901.181.560.470.160.04700.980.29–n/a–n/a262144qwen2–apiopenrouter
GPT-5.1openaiopenaiunknown0.630.185.001.411.720.480.06250.0176–n/a–n/a–n/a400000o200k–apiopenrouter
Gemini 2.5 Progooglegoogleunknown0.63n/a5.00n/a1.72n/a0.0625n/a0.19n/a–n/a–n/a1048576gemini–apiopenrouter
GPT-5.1openaiopenai–0.630.185.001.411.720.480.06250.0176–n/a–n/a–n/a400000o200k–apiopenrouter
Gemini 2.5 Progooglegoogle–0.63n/a5.00n/a1.72n/a0.0625n/a0.19n/a–n/a–n/a1048576gemini–apiopenrouter
Nova Pro (EU)amazonaws-bedrock–1.05n/a4.20n/a1.84n/a0.26n/a–n/a–n/a–n/a300000––aggregatorlitellm
o4-miniopenaiopenrouter–1.100.314.401.241.930.540.280.0775–n/a0.550.152.200.62200000o200k–apiopenrouter
o4-miniopenaiopenaiunknown1.100.314.401.241.930.540.280.0775–n/a–n/a–n/a200000o200k–apiopenrouter
o4-miniopenaiazure–1.100.314.401.241.930.540.280.0775–n/a–n/a–n/a200000o200k–aggregatorlitellm
o4-miniopenaiazure–1.100.314.401.241.930.540.280.0775–n/a0.550.152.200.62200000o200k–aggregatorlitellm
o4-miniopenaiopenai–1.100.314.401.241.930.540.280.0775–n/a0.550.152.200.62200000o200k–aggregatorlitellm
o4-miniopenaiopenai–1.100.314.401.241.930.540.280.0775–n/a0.550.152.200.62200000o200k–aggregatorlitellm
o4-miniopenaiopenai–1.100.314.401.241.930.540.280.0775–n/a–n/a–n/a200000o200k–apiopenrouter
Claude Haiku 4.5 (latest)anthropicopenrouter–1.00n/a5.00n/a2.00n/a0.10n/a1.25n/a0.50n/a2.50n/a200000claude–apiopenrouter
GPT-6.1 Sol (Global)openaiopenaiunknown1.000.285.001.412.000.560.05000.01411.250.35–n/a–n/a1050000o200k–apiopenrouter
Claude Haiku 4.5 (latest)anthropicazureunknown1.00n/a5.00n/a2.00n/a0.10n/a1.25n/a–n/a–n/a200000claude–apiopenrouter
Claude Haiku 4.5 (latest)anthropicaws-bedrockunknown1.00n/a5.00n/a2.00n/a0.10n/a1.25n/a–n/a–n/a200000claude–apiopenrouter
Claude Haiku 4.5 (latest)anthropicgoogle-vertexunknown1.00n/a5.00n/a2.00n/a0.10n/a1.25n/a–n/a–n/a200000claude–apiopenrouter
Claude Haiku 4.5 (latest)anthropicanthropicunknown1.00n/a5.00n/a2.00n/a0.10n/a1.25n/a–n/a–n/a200000claude–apiopenrouter
Claude Haiku 4.5 (latest)anthropicaws-bedrock–1.00n/a5.00n/a2.00n/a0.10n/a1.25n/a0.50n/a2.50n/a200000claude–aggregatorlitellm
Claude Haiku 4.5 (latest)anthropicaws-bedrock–1.00n/a5.00n/a2.00n/a0.10n/a1.25n/a–n/a–n/a200000claude–aggregatorlitellm
Claude Haiku 4.5 (latest)anthropicazure–1.00n/a5.00n/a2.00n/a0.10n/a1.25n/a–n/a–n/a200000claude–aggregatorlitellm
Claude Haiku 4.5 (latest)anthropicanthropic–1.00n/a5.00n/a2.00n/a0.10n/a1.25n/a0.50n/a2.50n/a200000claude–aggregatorlitellm
Claude Haiku 4.5 (latest)anthropicanthropic–1.00n/a5.00n/a2.00n/a0.10n/a1.25n/a0.50n/a2.50n/a200000claude–officialofficial-anthropic
Claude Haiku 4.5 (latest)anthropicaws-bedrock–1.00n/a5.00n/a2.00n/a0.10n/a1.25n/a0.50n/a2.50n/a200000claude–aggregatorlitellm
Qwen3 235B A22B Instruct 2507alibabahyperbolic–2.000.602.000.602.000.60–n/a–n/a–n/a–n/a131072qwen2–aggregatorlitellm
Kimi K2 Instructmoonshothyperbolic–2.00n/a2.00n/a2.00n/a–n/a–n/a–n/a–n/a262144––aggregatorlitellm
Claude Haiku 4.5 (latest)anthropicperplexity–1.00n/a5.00n/a2.00n/a0.10n/a–n/a–n/a–n/a200000claude–aggregatorlitellm
Claude Haiku 4.5 (latest)anthropicgoogle-vertex–1.00n/a5.00n/a2.00n/a0.10n/a1.25n/a0.50n/a2.50n/a200000claude–aggregatorlitellm
Claude Haiku 4.5 (latest)anthropicgoogle-vertex–1.00n/a5.00n/a2.00n/a0.10n/a1.25n/a0.50n/a2.50n/a200000claude–aggregatorlitellm
Claude Haiku 4.5 (latest)anthropicdeepinfra–1.00n/a5.00n/a2.00n/a–n/a–n/a–n/a–n/a200000claude–aggregatorlitellm
GPT-6.1 Sol (Global)openaiopenaibatch1.000.285.001.412.000.560.05000.01411.250.35–n/a–n/a1050000o200k–officialofficial-openai
GPT-6.1 Sol (Global)openaiopenaiflex1.000.285.001.412.000.560.05000.01411.250.35–n/a–n/a1050000o200k–officialofficial-openai
GPT-6.1 Sol (Global)openaiopenai–1.000.285.001.412.000.560.05000.01411.250.35–n/a–n/a1050000o200k–apiopenrouter
Claude Haiku 4.5 (latest)anthropicazure–1.00n/a5.00n/a2.00n/a0.10n/a1.25n/a–n/a–n/a200000claude–apiopenrouter
Claude Haiku 4.5 (latest)anthropicaws-bedrock–1.00n/a5.00n/a2.00n/a0.10n/a1.25n/a–n/a–n/a200000claude–apiopenrouter
Claude Haiku 4.5 (latest)anthropicgoogle-vertex–1.00n/a5.00n/a2.00n/a0.10n/a1.25n/a–n/a–n/a200000claude–apiopenrouter

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