api cost calculator
LLM API cost calculator
Estimate and compare what your workload costs per month across 298+ models. Set the tokens and request volume; the verdict and table update live.
cheapest for this workload
Ling-2.6-flash
$6.00 / mo
vs o1-pro at $105,000/mo — 17500× more
| model | provider | input /1M | output /1M | monthly cost ↑ |
|---|---|---|---|---|
| Ling-2.6-flash routed | Inclusionai | $0.01 | $0.03 | $6.00 |
| Mistral Nemo routed | Mistral | $0.02 | $0.03 | $9.00 |
| Llama 3.1 8B Instruct routed | Meta | $0.02 | $0.03 | $9.00 |
| Granite 4.0 Micro routed | Ibm Granite | $0.02 | $0.11 | $16.30 |
| Llama 3 8B Lunaris routed | Sao10k | $0.04 | $0.05 | $17.00 |
| LFM2-24B-A2B routed | Liquid | $0.03 | $0.12 | $21.00 |
| Qwen2.5 7B Instruct routed | Qwen | $0.04 | $0.10 | $22.00 |
| gpt-oss-20b routed | OpenAI | $0.03 | $0.14 | $22.70 |
| Mistral Small 3 routed | Mistral | $0.05 | $0.08 | $23.00 |
| MythoMax 13B routed | Gryphe | $0.06 | $0.06 | $24.00 |
| Nova Micro 1.0 routed | Amazon | $0.04 | $0.14 | $24.50 |
| Granite 4.1 8B routed | Ibm Granite | $0.05 | $0.10 | $25.00 |
| Gemma 3 4B routed | $0.05 | $0.10 | $25.00 | |
| Command R7B (12-2024) routed | Cohere | $0.04 | $0.15 | $26.25 |
| Llama 3.2 1B Instruct routed | Meta | $0.03 | $0.20 | $28.20 |
| Trinity Mini routed | Arcee Ai | $0.04 | $0.15 | $28.50 |
| gpt-oss-120b routed | OpenAI | $0.04 | $0.18 | $29.70 |
| Gemma 3n 4B routed | $0.06 | $0.12 | $30.00 | |
| Gemma 3 12B routed | $0.05 | $0.15 | $30.00 | |
| Phi 4 routed | Microsoft | $0.07 | $0.14 | $33.50 |
| Qwen3 30B A3B Instruct 2507 routed | Qwen | $0.05 | $0.19 | $33.75 |
| Nemotron 3 Nano 30B A3B routed | NVIDIA | $0.05 | $0.20 | $35.00 |
| Qwen3 235B A22B Instruct 2507 routed | Qwen | $0.09 | $0.10 | $37.00 |
| Hy3 preview routed | Tencent | $0.06 | $0.21 | $39.90 |
| Reka Edge routed | Rekaai | $0.10 | $0.10 | $40.00 |
| Qwen3 235B A22B Thinking 2507 routed | Qwen | $0.10 | $0.10 | $40.00 |
| Ministral 3 3B 2512 routed | Mistral | $0.10 | $0.10 | $40.00 |
| Gemma 3 27B routed | $0.08 | $0.16 | $40.00 | |
| Nova Lite 1.0 routed | Amazon | $0.06 | $0.24 | $42.00 |
| Mistral Small 3.2 24B routed | Mistral | $0.07 | $0.20 | $42.50 |
| Qwen3.5-9B routed | Qwen | $0.10 | $0.15 | $45.00 |
| Qwen3.5-Flash routed | Qwen | $0.07 | $0.26 | $45.50 |
| Qwen3 Coder 30B A3B Instruct routed | Qwen | $0.07 | $0.27 | $48.00 |
| Llama 3.2 3B Instruct routed | Meta | $0.05 | $0.34 | $48.77 |
| UI-TARS 7B routed | ByteDance | $0.10 | $0.20 | $50.00 |
| Reka Flash 3 routed | Rekaai | $0.10 | $0.20 | $50.00 |
| Gemma 4 26B A4B routed | $0.06 | $0.33 | $51.00 | |
| Qwen3 32B routed | Qwen | $0.08 | $0.28 | $52.00 |
| Seed 1.6 Flash routed | Bytedance Seed | $0.07 | $0.30 | $52.50 |
| gpt-oss-safeguard-20b routed | OpenAI | $0.07 | $0.30 | $52.50 |
| Qwen3 14B routed | Qwen | $0.10 | $0.24 | $54.00 |
| Qwen3 8B routed | Qwen | $0.05 | $0.40 | $55.00 |
| GPT-5 Nano | OpenAI | $0.05 | $0.40 | $55.00 |
| Llama 3 8B Instruct routed | Meta | $0.14 | $0.14 | $56.00 |
| Step 3.5 Flash routed | Stepfun | $0.09 | $0.30 | $57.00 |
| GLM 4.7 Flash routed | Z.AI | $0.06 | $0.40 | $58.00 |
| Phi 4 Mini Instruct routed | Microsoft | $0.08 | $0.35 | $59.00 |
| Rnj 1 Instruct routed | Essentialai | $0.15 | $0.15 | $60.00 |
| Ministral 3 8B 2512 routed | Mistral | $0.15 | $0.15 | $60.00 |
| Voxtral Small 24B 2507 routed | Mistral | $0.10 | $0.30 | $60.00 |
| MiMo-V2-Flash routed | Xiaomi | $0.10 | $0.30 | $60.00 |
| Llama 4 Scout routed | Meta | $0.10 | $0.30 | $60.00 |
| Llama 3.3 70B Instruct routed | Meta | $0.10 | $0.32 | $62.00 |
| Qwen3 30B A3B Thinking 2507 routed | Qwen | $0.08 | $0.40 | $64.00 |
| Seed-2.0-Mini routed | Bytedance Seed | $0.10 | $0.40 | $70.00 |
| MiMo-V2.5 routed | Xiaomi | $0.14 | $0.28 | $70.00 |
| GPT-4.1 Nano | OpenAI | $0.10 | $0.40 | $70.00 |
| Gemini 2.5 Flash-Lite | $0.10 | $0.40 | $70.00 | |
| Gemini 2.5 Flash Lite Preview 09-2025 | $0.10 | $0.40 | $70.00 | |
| DeepSeek V4 Flash first-party | DeepSeek | $0.14 | $0.28 | $70.00 |
| Gemma 4 31B routed | $0.12 | $0.35 | $71.00 | |
| Nemotron 3 Super routed | NVIDIA | $0.09 | $0.45 | $72.00 |
| Llama Guard 4 12B routed | Meta | $0.18 | $0.18 | $72.00 |
| Qwen3 VL 32B Instruct routed | Qwen | $0.10 | $0.42 | $72.80 |
| Qwen3 VL 8B Instruct routed | Qwen | $0.08 | $0.50 | $74.00 |
| Hermes 4 70B routed | Nousresearch | $0.13 | $0.40 | $79.00 |
| Ministral 3 14B 2512 routed | Mistral | $0.20 | $0.20 | $80.00 |
| Ring-2.6-1T routed | Inclusionai | $0.07 | $0.63 | $85.00 |
| Ling-2.6-1T routed | Inclusionai | $0.07 | $0.63 | $85.00 |
| Qwen3 30B A3B routed | Qwen | $0.12 | $0.50 | $86.00 |
| Qwen3 VL 30B A3B Instruct routed | Qwen | $0.13 | $0.52 | $91.00 |
| Rocinante 12B routed | Thedrummer | $0.17 | $0.43 | $94.00 |
| Olmo 3 32B Think routed | Allenai | $0.15 | $0.50 | $95.00 |
| Hunyuan A13B Instruct routed | Tencent | $0.14 | $0.57 | $99.00 |
| DeepSeek V3.2 routed | DeepSeek | $0.23 | $0.34 | $103 |
| Solar Pro 3 routed | Upstage | $0.15 | $0.60 | $105 |
| Mistral Small 4 routed | Mistral | $0.15 | $0.60 | $105 |
| Llama 4 Maverick routed | Meta | $0.15 | $0.60 | $105 |
| GPT-4o-mini Search Preview | OpenAI | $0.15 | $0.60 | $105 |
| GPT-4o-mini (2024-07-18) | OpenAI | $0.15 | $0.60 | $105 |
| GPT-4o-mini | OpenAI | $0.15 | $0.60 | $105 |
| Command R (08-2024) routed | Cohere | $0.15 | $0.60 | $105 |
| Qwen3 Next 80B A3B Thinking routed | Qwen | $0.10 | $0.78 | $107 |
| Qwen3 Coder Next routed | Qwen | $0.11 | $0.80 | $113 |
| R1 Distill Qwen 32B routed | DeepSeek | $0.29 | $0.29 | $116 |
| Saba routed | Mistral | $0.20 | $0.60 | $120 |
| DeepSeek V3.2 Exp routed | DeepSeek | $0.27 | $0.41 | $122 |
| GLM 4.5 Air routed | Z.AI | $0.13 | $0.85 | $123 |
| MiniMax M2.5 routed | MiniMax | $0.15 | $0.90 | $135 |
| Qwen3 Next 80B A3B Instruct routed | Qwen | $0.09 | $1.10 | $137 |
| DeepSeek V3 0324 routed | DeepSeek | $0.20 | $0.77 | $137 |
| Llama 3.2 11B Vision Instruct routed | Meta | $0.34 | $0.34 | $138 |
| Cydonia 24B V4.1 routed | Thedrummer | $0.30 | $0.50 | $140 |
| DeepSeek V3 routed | DeepSeek | $0.20 | $0.80 | $140 |
| Qwen3.5-35B-A3B routed | Qwen | $0.14 | $1.00 | $142 |
| DeepSeek V3.1 routed | DeepSeek | $0.21 | $0.79 | $142 |
| Qwen3.6 35B A3B routed | Qwen | $0.15 | $1.00 | $145 |
| Qwen3 VL 235B A22B Instruct routed | Qwen | $0.20 | $0.88 | $148 |
| Qwen2.5 72B Instruct routed | Qwen | $0.36 | $0.40 | $148 |
| Mercury 2 routed | Inception | $0.25 | $0.75 | $150 |
| Trinity Large Thinking routed | Arcee Ai | $0.22 | $0.85 | $151 |
| Qwen3 Coder Flash routed | Qwen | $0.20 | $0.97 | $156 |
| Qwen-Plus routed | Qwen | $0.26 | $0.78 | $156 |
| Qwen Plus 0728 (thinking) routed | Qwen | $0.26 | $0.78 | $156 |
| Qwen Plus 0728 routed | Qwen | $0.26 | $0.78 | $156 |
| UnslopNemo 12B routed | Thedrummer | $0.40 | $0.40 | $160 |
| Llama 3.3 Nemotron Super 49B V1.5 routed | NVIDIA | $0.40 | $0.40 | $160 |
| Llama 3.1 70B Instruct routed | Meta | $0.40 | $0.40 | $160 |
| Mistral Small 3.1 24B routed | Mistral | $0.35 | $0.56 | $161 |
| Qwen3.6 Flash routed | Qwen | $0.19 | $1.13 | $169 |
| MiniMax-01 routed | MiniMax | $0.20 | $1.10 | $170 |
| INTELLECT-3 routed | Prime Intellect | $0.20 | $1.10 | $170 |
| Qwen3 VL 8B Thinking routed | Qwen | $0.12 | $1.36 | $172 |
| Step 3.7 Flash routed | Stepfun | $0.20 | $1.15 | $175 |
| MiniMax M2.7 routed | MiniMax | $0.25 | $1.00 | $175 |
| DeepSeek V3.1 Terminus routed | DeepSeek | $0.27 | $0.95 | $176 |
| MiniMax M2 routed | MiniMax | $0.26 | $1.00 | $177 |
| GLM 4.6V routed | Z.AI | $0.30 | $0.90 | $180 |
| Codestral 2508 routed | Mistral | $0.30 | $0.90 | $180 |
| MiniMax M2.1 routed | MiniMax | $0.29 | $0.95 | $182 |
| GPT-5.4 nano | OpenAI | $0.20 | $1.25 | $185 |
| Perceptron Mk1 routed | Perceptron | $0.15 | $1.50 | $195 |
| Qwen3 VL 30B A3B Thinking routed | Qwen | $0.13 | $1.56 | $195 |
| ReMM SLERP 13B routed | Undi95 | $0.45 | $0.65 | $200 |
| Claude 3 Haiku | Anthropic | $0.25 | $1.25 | $200 |
| MiniMax M3 routed | MiniMax | $0.30 | $1.20 | $210 |
| MiniMax M2-her routed | MiniMax | $0.30 | $1.20 | $210 |
| KAT-Coder-Pro V2 routed | Kwaipilot | $0.30 | $1.20 | $210 |
| Qwen3.5-27B routed | Qwen | $0.20 | $1.56 | $215 |
| MiMo-V2.5-Pro routed | Xiaomi | $0.43 | $0.87 | $217 |
| DeepSeek V4 Pro | DeepSeek | $0.43 | $0.87 | $217 |
| Qwen3.7 Plus routed | Qwen | $0.32 | $1.28 | $224 |
| Gemini 3.1 Flash Lite Preview | $0.25 | $1.50 | $225 | |
| Gemini 3.1 Flash Lite | $0.25 | $1.50 | $225 | |
| Llama 3 70B Instruct routed | Meta | $0.51 | $0.74 | $227 |
| Coder Large routed | Arcee Ai | $0.50 | $0.80 | $230 |
| Qwen3.5 Plus 2026-02-15 routed | Qwen | $0.26 | $1.56 | $234 |
| Skyfall 36B V2 routed | Thedrummer | $0.55 | $0.80 | $245 |
| Qwen3 Coder 480B A35B routed | Qwen | $0.22 | $1.80 | $246 |
| WizardLM-2 8x22B routed | Microsoft | $0.62 | $0.62 | $248 |
| ERNIE 4.5 VL 424B A47B routed | Baidu | $0.42 | $1.25 | $251 |
| Gemma 2 27B routed | $0.65 | $0.65 | $260 | |
| Qwen3.5 Plus 2026-04-20 routed | Qwen | $0.30 | $1.80 | $270 |
| Llama 3.3 Euryale 70B routed | Sao10k | $0.65 | $0.75 | $270 |
| Seed-2.0-Lite routed | Bytedance Seed | $0.25 | $2.00 | $275 |
| Seed 1.6 routed | Bytedance Seed | $0.25 | $2.00 | $275 |
| GPT-5.1-Codex-Mini | OpenAI | $0.25 | $2.00 | $275 |
| GPT-5 Mini | OpenAI | $0.25 | $2.00 | $275 |
| Hermes 3 70B Instruct routed | Nousresearch | $0.70 | $0.70 | $280 |
| GPT-4.1 Mini | OpenAI | $0.40 | $1.60 | $280 |
| Qwen3.5-122B-A10B routed | Qwen | $0.26 | $2.08 | $286 |
| Qwen3.6 Plus routed | Qwen | $0.33 | $1.95 | $293 |
| GLM 4.7 routed | Z.AI | $0.40 | $1.75 | $295 |
| Qwen2.5 Coder 32B Instruct routed | Qwen | $0.66 | $1.00 | $298 |
| Mistral Large 3 | Mistral | $0.50 | $1.50 | $300 |
| GPT-3.5 Turbo | OpenAI | $0.50 | $1.50 | $300 |
| GLM 4.6 routed | Z.AI | $0.43 | $1.74 | $303 |
| Kimi K2.5 routed | Moonshot | $0.38 | $2.02 | $315 |
| Qwen3 235B A22B routed | Qwen | $0.46 | $1.82 | $319 |
| R1 Distill Llama 70B routed | DeepSeek | $0.80 | $0.80 | $320 |
| Mistral Medium 3.1 routed | Mistral | $0.40 | $2.00 | $320 |
| Mistral Medium 3 routed | Mistral | $0.40 | $2.00 | $320 |
| Devstral 2 2512 routed | Mistral | $0.40 | $2.00 | $320 |
| Weaver (alpha) routed | Mancer | $0.75 | $1.00 | $325 |
| Qwen3 VL 235B A22B Thinking routed | Qwen | $0.26 | $2.60 | $338 |
| Qwen2.5 VL 72B Instruct routed | Qwen | $0.80 | $1.00 | $340 |
| Nova 2 Lite routed | Amazon | $0.30 | $2.50 | $340 |
| Nano Banana (Gemini 2.5 Flash Image) | $0.30 | $2.50 | $340 | |
| MiniMax M1 routed | MiniMax | $0.40 | $2.20 | $340 |
| Llama 3.1 Euryale 70B v2.2 routed | Sao10k | $0.85 | $0.85 | $340 |
| Gemini 2.5 Flash | $0.30 | $2.50 | $340 | |
| Virtuoso Large routed | Arcee Ai | $0.75 | $1.20 | $345 |
| Aion-1.0-Mini routed | Aion Labs | $0.70 | $1.40 | $350 |
| Qwen3.5 397B A17B routed | Qwen | $0.39 | $2.34 | $351 |
| Morph V3 Fast routed | Morph | $0.80 | $1.20 | $360 |
| GLM 4.5V routed | Z.AI | $0.60 | $1.80 | $360 |
| R1 0528 routed | DeepSeek | $0.50 | $2.15 | $365 |
| GLM 5 routed | Z.AI | $0.60 | $1.92 | $372 |
| Relace Apply 3 routed | Relace | $0.85 | $1.25 | $380 |
| Sonar routed | Perplexity | $1.00 | $1.00 | $400 |
| Nemotron 3 Ultra routed | NVIDIA | $0.50 | $2.50 | $400 |
| Hermes 3 405B Instruct routed | Nousresearch | $1.00 | $1.00 | $400 |
| GLM 4.5 routed | Z.AI | $0.60 | $2.20 | $400 |
| Aion-RP 1.0 (8B) routed | Aion Labs | $0.80 | $1.60 | $400 |
| Aion-2.0 routed | Aion Labs | $0.80 | $1.60 | $400 |
| Kimi K2 0711 routed | Moonshot | $0.57 | $2.30 | $401 |
| Qwen3.6 27B routed | Qwen | $0.29 | $3.17 | $404 |
| GPT Audio Mini | OpenAI | $0.60 | $2.40 | $420 |
| Kimi K2 Thinking routed | Moonshot | $0.60 | $2.50 | $430 |
| Kimi K2 0905 routed | Moonshot | $0.60 | $2.50 | $430 |
| Nano Banana 2 (Gemini 3.1 Flash Image Preview) | $0.50 | $3.00 | $450 | |
| Gemini 3 Flash Preview | $0.50 | $3.00 | $450 | |
| R1 routed | DeepSeek | $0.70 | $2.50 | $460 |
| Morph V3 Large routed | Morph | $0.90 | $1.90 | $460 |
| Grok Build 0.1 | xAI | $1.00 | $2.00 | $500 |
| GPT-3.5 Turbo (older v0613) | OpenAI | $1.00 | $2.00 | $500 |
| Cogito v2.1 671B routed | Deepcogito | $1.25 | $1.25 | $500 |
| Qwen3 Coder Plus routed | Qwen | $0.65 | $3.25 | $520 |
| Kimi K2.6 routed | Moonshot | $0.68 | $3.41 | $545 |
| Nova Pro 1.0 routed | Amazon | $0.80 | $3.20 | $560 |
| Kimi K2.7 Code routed | Moonshot | $0.75 | $3.50 | $575 |
| Switchpoint Router routed | Switchpoint | $0.85 | $3.40 | $595 |
| Relace Search routed | Relace | $1.00 | $3.00 | $600 |
| Hermes 4 405B routed | Nousresearch | $1.00 | $3.00 | $600 |
| GLM 5.1 routed | Z.AI | $0.98 | $3.08 | $602 |
| Qwen3 Max Thinking routed | Qwen | $0.78 | $3.90 | $624 |
| Qwen3 Max routed | Qwen | $0.78 | $3.90 | $624 |
| Grok 4.3 | xAI | $1.25 | $2.50 | $625 |
| Grok 4.20 | xAI | $1.25 | $2.50 | $625 |
| Claude 3.5 Haiku | Anthropic | $0.80 | $4.00 | $640 |
| GPT-3.5 Turbo Instruct | OpenAI | $1.50 | $2.00 | $650 |
| GPT-5.4 mini | OpenAI | $0.75 | $4.50 | $675 |
| Qwen3.7 Max routed | Qwen | $1.25 | $3.75 | $750 |
| GLM 5 Turbo routed | Z.AI | $1.20 | $4.00 | $760 |
| o4 Mini High | OpenAI | $1.10 | $4.40 | $770 |
| o4 Mini | OpenAI | $1.10 | $4.40 | $770 |
| o3 Mini High | OpenAI | $1.10 | $4.40 | $770 |
| o3 Mini | OpenAI | $1.10 | $4.40 | $770 |
| Palmyra X5 routed | Writer | $0.60 | $6.00 | $780 |
| Claude Haiku 4.5 | Anthropic | $1.00 | $5.00 | $800 |
| Qwen3.6 Max Preview routed | Qwen | $1.04 | $6.24 | $936 |
| GPT-5 Image Mini | OpenAI | $2.50 | $2.00 | $950 |
| Mixtral 8x22B Instruct routed | Mistral | $2.00 | $6.00 | $1,200 |
| Mistral Medium 3.5 | Mistral | $1.50 | $7.50 | $1,200 |
| Mistral Large 2407 routed | Mistral | $2.00 | $6.00 | $1,200 |
| Mistral Large routed | Mistral | $2.00 | $6.00 | $1,200 |
| Llama 3.1 70B Hanami x1 routed | Sao10k | $3.00 | $3.00 | $1,200 |
| Grok 4.20 Multi-Agent | xAI | $2.00 | $6.00 | $1,200 |
| GPT-3.5 Turbo 16k | OpenAI | $3.00 | $4.00 | $1,300 |
| Gemini 3.5 Flash | $1.50 | $9.00 | $1,350 | |
| GPT-5.1-Codex-Max | OpenAI | $1.25 | $10.00 | $1,375 |
| GPT-5.1-Codex | OpenAI | $1.25 | $10.00 | $1,375 |
| GPT-5.1 Chat | OpenAI | $1.25 | $10.00 | $1,375 |
| GPT-5.1 | OpenAI | $1.25 | $10.00 | $1,375 |
| GPT-5 Codex | OpenAI | $1.25 | $10.00 | $1,375 |
| GPT-5 Chat | OpenAI | $1.25 | $10.00 | $1,375 |
| GPT-5 | OpenAI | $1.25 | $10.00 | $1,375 |
| Gemini 2.5 Pro Preview 06-05 | $1.25 | $10.00 | $1,375 | |
| Gemini 2.5 Pro Preview 05-06 | $1.25 | $10.00 | $1,375 | |
| Gemini 2.5 Pro | $1.25 | $10.00 | $1,375 | |
| Sonar Reasoning Pro routed | Perplexity | $2.00 | $8.00 | $1,400 |
| Sonar Deep Research routed | Perplexity | $2.00 | $8.00 | $1,400 |
| o4 Mini Deep Research | OpenAI | $2.00 | $8.00 | $1,400 |
| o3 | OpenAI | $2.00 | $8.00 | $1,400 |
| Jamba Large 1.7 routed | AI21 | $2.00 | $8.00 | $1,400 |
| GPT-4.1 | OpenAI | $2.00 | $8.00 | $1,400 |
| Magnum v4 72B routed | Anthracite Org | $3.00 | $5.00 | $1,400 |
| Inflection 3 Productivity routed | Inflection | $2.50 | $10.00 | $1,750 |
| Inflection 3 Pi routed | Inflection | $2.50 | $10.00 | $1,750 |
| GPT-4o Search Preview | OpenAI | $2.50 | $10.00 | $1,750 |
| GPT-4o (2024-11-20) | OpenAI | $2.50 | $10.00 | $1,750 |
| GPT-4o (2024-08-06) | OpenAI | $2.50 | $10.00 | $1,750 |
| GPT-4o | OpenAI | $2.50 | $10.00 | $1,750 |
| GPT Audio | OpenAI | $2.50 | $10.00 | $1,750 |
| Command R+ (08-2024) routed | Cohere | $2.50 | $10.00 | $1,750 |
| Command A routed | Cohere | $2.50 | $10.00 | $1,750 |
| Nano Banana Pro (Gemini 3 Pro Image Preview) | $2.00 | $12.00 | $1,800 | |
| Gemini 3.1 Pro Preview Custom Tools | $2.00 | $12.00 | $1,800 | |
| Gemini 3.1 Pro | $2.00 | $12.00 | $1,800 | |
| GPT-5.3-Codex | OpenAI | $1.75 | $14.00 | $1,925 |
| GPT-5.3 Chat | OpenAI | $1.75 | $14.00 | $1,925 |
| GPT-5.2-Codex | OpenAI | $1.75 | $14.00 | $1,925 |
| GPT-5.2 Chat | OpenAI | $1.75 | $14.00 | $1,925 |
| GPT-5.2 | OpenAI | $1.75 | $14.00 | $1,925 |
| Nova Premier 1.0 routed | Amazon | $2.50 | $12.50 | $2,000 |
| Aion-1.0 routed | Aion Labs | $4.00 | $8.00 | $2,000 |
| GPT-5.4 | OpenAI | $2.50 | $15.00 | $2,250 |
| Sonar Pro Search routed | Perplexity | $3.00 | $15.00 | $2,400 |
| Sonar Pro routed | Perplexity | $3.00 | $15.00 | $2,400 |
| Claude Sonnet 4.6 | Anthropic | $3.00 | $15.00 | $2,400 |
| Claude Sonnet 4.5 | Anthropic | $3.00 | $15.00 | $2,400 |
| Claude Sonnet 4 | Anthropic | $3.00 | $15.00 | $2,400 |
| GPT-4o (2024-05-13) | OpenAI | $5.00 | $15.00 | $3,000 |
| GPT-5.4 Image 2 | OpenAI | $8.00 | $15.00 | $3,900 |
| GPT-5 Image | OpenAI | $10.00 | $10.00 | $4,000 |
| Claude Opus 4.8 | Anthropic | $5.00 | $25.00 | $4,000 |
| Claude Opus 4.7 | Anthropic | $5.00 | $25.00 | $4,000 |
| Claude Opus 4.6 | Anthropic | $5.00 | $25.00 | $4,000 |
| Claude Opus 4.5 | Anthropic | $5.00 | $25.00 | $4,000 |
| GPT-5.5 | OpenAI | $5.00 | $30.00 | $4,500 |
| GPT Chat Latest | OpenAI | $5.00 | $30.00 | $4,500 |
| GPT-4 Turbo Preview | OpenAI | $10.00 | $30.00 | $6,000 |
| GPT-4 Turbo | OpenAI | $10.00 | $30.00 | $6,000 |
| o3 Deep Research | OpenAI | $10.00 | $40.00 | $7,000 |
| Claude Opus 4.8 (Fast) | Anthropic | $10.00 | $50.00 | $8,000 |
| Claude Fable 5 | Anthropic | $10.00 | $50.00 | $8,000 |
| o1 | OpenAI | $15.00 | $60.00 | $10,500 |
| Claude Opus 4.1 | Anthropic | $15.00 | $75.00 | $12,000 |
| Claude Opus 4 | Anthropic | $15.00 | $75.00 | $12,000 |
| o3 Pro | OpenAI | $20.00 | $80.00 | $14,000 |
| GPT-4 | OpenAI | $30.00 | $60.00 | $15,000 |
| GPT-5 Pro | OpenAI | $15.00 | $120 | $16,500 |
| GPT-5.2 Pro | OpenAI | $21.00 | $168 | $23,100 |
| Claude Opus 4.7 (Fast) | Anthropic | $30.00 | $150 | $24,000 |
| Claude Opus 4.6 (Fast) | Anthropic | $30.00 | $150 | $24,000 |
| GPT-5.5 Pro | OpenAI | $30.00 | $180 | $27,000 |
| GPT-5.4 Pro | OpenAI | $30.00 | $180 | $27,000 |
| o1-pro | OpenAI | $150 | $600 | $105,000 |
List prices for standard usage; snapshot . How we measure · all model specs →
How the calculator works
For each model: (input tokens ÷ 1M × input price) + (output tokens ÷ 1M × output price), times your requests per month. The interesting part is what it reveals once both token directions are in play.
Why output tokens usually run the bill
Output is priced 3–5× higher than input across every provider. So the ratio of output to input in your workload often matters more than either headline price. A summariser is input-bound; a code generator or chat assistant is output-bound, and the cheapest model flips accordingly. Change the token fields and watch the ranking reorder.
Routed vs first-party prices
Most rows show OpenRouter's routed market price (routed); premier providers also carry their official list price ( / first-party). When a reseller routes a model cheaper than the lab's own API, we keep the first-party price and tag it — see methodology.
What this doesn't include
- Batch discounts (~50% on async) and cached-input rates (~90% off reused prefixes).
- Tiered long-context rates above a prompt-size threshold; figures use the standard tier.
- Fine-tuning, image/audio surcharges, and request overhead beyond the tokens you enter.
These are list/market prices, not measured outcomes — two models at the same rate can cost very different amounts to finish a task. That gap is why we also publish measured cost-per-task.
Frequently asked questions
Which LLM API is cheapest?
It depends on your workload. A low input price can still lose once output volume is counted, because output tokens cost several times more. Enter your real tokens-in, tokens-out and requests above — the cheapest headline rate is rarely the cheapest finished job.
What do the 'routed' and 'first-party' tags mean?
Most models are priced from OpenRouter's routed market price. The premier providers (OpenAI, Anthropic, Google, xAI, DeepSeek, Mistral) also carry the official first-party list price; where the two differ, the first-party value wins and the row is tagged first-party. See the methodology page.
Do reasoning or 'thinking' tokens cost extra?
Yes — they're billed as output even when hidden. A low headline rate can cost more per finished task if the model reasons at length. Fold typical thinking tokens into the output field, or see measured cost-per-task on the real-cost index.
How current are these prices?
Captured from OpenRouter and official sources with a date; latest snapshot 2026-06-15, refreshed every other day. Treat as a sourced reference and confirm with the provider before relying on it.
Related
- Token counter — how many tokens your text is, per model.
- Model comparison — context windows, sources, prices side by side.
- LLM pricing explained — input vs output vs cached vs reasoning tokens.