Skip to content

Cheapest models

Ranked by blended API price (75% input + 25% output, USD per 1M tokens). Free promotional listings excluded.

Ranked by blended API price (75% input + 25% output, USD per 1M tokens). Free promotional listings excluded. Source: OpenRouter. Variants (reasoning effort, batch) are folded into one row.

#ModelCapabilityReasoningCodingContext$/M
201Aion-2.0AionLabs———131K$1.00
202Aion-RP 1.0 (8B)AionLabs———33K$1.00
203Claude Haiku 4.5batchAnthropic · +1 variant———200K$1.00
204GLM 4.5Z.ai (Zhipu AI)———131K$1.00
205Hermes 3 405B InstructNous Research———131K$1.00
206Seed 2.1 TurboByteDance———262K$1.00
207SonarPerplexity———127K$1.00
208Kimi K2 0711Moonshot AI———131K$1.00
209Qwen3.6 27BAlibaba (Qwen)146.585.9—262K$1.04
210Nemotron 3 UltraNVIDIA146.285.4—262K$1.05
211GPT Audio MiniOpenAI———128K$1.05
212Kimi K2 ThinkingMoonshot AI146.0——262K$1.07
213Kimi K2 0905Moonshot AI———262K$1.07
214Nano Banana 2 (Gemini 3.1 Flash Image Preview)Google———66K$1.13
215Nano Banana 2 (Gemini 3.1 Flash Image)Google———131K$1.13
216Seed-2.0-CodeByteDance———262K$1.13
217Qwen3.8 27BAlibaba (Qwen)———1M$1.14
218R1DeepSeek139.071.7—64K$1.15
219Morph V3 LargeMorph———262K$1.15
220DeepSeek V4 Pro 0423DeepSeek149.1——1.05M$1.19
221GPT-3.5 Turbo (older v0613)OpenAI———4K$1.25
222Grok 4.3batchxAI · +1 variant———1M$1.25
223Grok Build 0.1xAI———256K$1.25
224Hy4 previewTencent———1.05M$1.25
225Qwen3.5 397B A17BAlibaba (Qwen)146.686.4—262K$1.29
226KAT-Coder-Pro V2.5Kuaishou———262K$1.29
227Qwen3 Coder PlusAlibaba (Qwen)———1M$1.30
228Qwen3 VL 235B A22B ThinkingAlibaba (Qwen)———131K$1.30
229Kimi K2.7 CodeMoonshot AI150.087.930.5262K$1.32
230Kimi K2.6Moonshot AI151.090.8—262K$1.34
231GLM 5.2Z.ai (Zhipu AI)151.891.943.81.05M$1.39
232Nova Pro 1.0Amazon———300K$1.40
233DeepSeek V4 Pro 0813DeepSeek155.491.7—1.05M$1.41
234Hermes 4 405BNous Research———131K$1.50
235Mistral Medium 3.5batchMistral AI · +1 variant———262K$1.50
236Relace SearchRelace———256K$1.50
237Qwen3 MaxAlibaba (Qwen)142.4——262K$1.56
238Qwen3 Max ThinkingAlibaba (Qwen)———262K$1.56
239Grok 4.20xAI152.089.3—2M$1.56
240Grok 4.20 Multi-AgentxAI———2M$1.56
241GPT-3.5 Turbo InstructOpenAI———4K$1.63
242Gemini 3.5 FlashbatchGoogle · +1 variant———1.05M$1.69
243Sakana NamazuSakana AI———262K$1.71
244Gemini 2.5 ProbatchGoogle · +1 variant———1.05M$1.72
245GPT-5batchOpenAI · +1 variant———400K$1.72
246GPT-5.1batchOpenAI · +1 variant———400K$1.72
247GPT-4.1batchOpenAI · +1 variant———1.05M$1.75
248o3batchOpenAI · +1 variant———200K$1.75
249InklingThinking Machines Lab148.6——524K$1.76
250GLM 5 TurboZ.ai (Zhipu AI)———203K$1.90
— means no published score from that source yet · click a model for every benchmark with its source

New - awaiting independent evaluation

All new models →

Not ranked until an independent evaluator (Epoch AI) publishes scores. We don't use vendor-reported benchmark claims for rankings.

Best AI for coding →
Models ranked on DeepSWE
Best AI for reasoning →
Ranked on GPQA Diamond
Best AI for math →
Ranked on competition math
Best open-weight models →
Weights you can run yourself
Cheapest capable models →
Lowest blended API price
Best AI coding tools →
Apps & agents by popularity
Best AI image generators →
Apps & models
Best AI for research →
Answer engines & deep research