Reasoning models
296 models that reason step by step before they answer, for math, code and multi-step problems.
/
| OpenAI GPT-5.4 MiniOpenAIPrompt cachingJSON+3 | $3.60/1Mvia APIMart | 400K | $0.75/1M | $4.50/1M | 1412 | — | |
| OpenAI GPT-5.4 NanoOpenAIPrompt cachingJSON+3 | $1/1Mvia APIMart | 400K | $0.20/1M | $1.25/1M | 1372 | — | |
| OpenAI GPT-5.4 ProOpenAIJSONReasoning+2 | $144/1Mvia APIMart | 1.05M | $30/1M | $180/1M | — | — | |
| OpenAI GPT-5.5OpenAIPrompt cachingJSON+3 | $24/1Mvia APIMart | 1.05M | $5/1M | $30/1M | 1513 | — | |
| OpenAI o1OpenAIPrompt cachingJSON+3 | $48/1Mvia APIMart | 200K | $15/1M | $60/1M | 1366 | — | |
| OpenAI o3OpenAIPrompt cachingJSON+3 | $6.40/1Mvia APIMart | 200K | $2/1M | $8/1M | 1410 | — | |
| OpenAI o3 miniOpenAIPrompt cachingJSON+2 | $3.52/1Mvia APIMart | 200K | $1.10/1M | $4.40/1M | 1337 | — | |
| Perceptron Mk1PerceptronJSONReasoning+1 | $1.50/1Mvia OpenRouter | 33K | $0.15/1M | $1.50/1M | — | — | |
| Perceptron Mk1.5PerceptronJSONReasoning+2 | $1.50/1Mvia OpenRouter | 37K | $0.15/1M | $1.50/1M | — | — | |
 Perplexity Sonar Deep ResearchPerplexityReasoning | $8/1Mvia FastRouter | 128K | $2/1M | $8/1M | — | — | |
 Qwen MaxAlibabaPrompt cachingJSON+2 | $1.38/1Mvia Alibaba | 33K | $0.345/1M | $1.38/1M | 1282 | — | |
| Qwen3 235B A22BAlibabaApache 2.0JSON+2 | $0.60/1Mvia Hugging Face | 128K | $0.20/1M | $0.60/1M | — | — | |
| Qwen3 235B A22b Thinking 2507AlibabaApache 2.0JSON+2 | $2.30/1Mvia Hugging Face | 262K | $0.23/1M | $2.30/1M | — | — | |
| Qwen3 MaxAlibabaPrompt cachingJSON+2 | $1.43/1Mvia Ofox | 262K | $0.36/1M | $1.43/1M | 1413 | — | |
| Qwen3 Max PreviewAlibabaPrompt cachingJSON+3 | $3.38/1Mvia AIHubMix | 262K | $0.846/1M | $3.38/1M | 1439 | — | |
| Qwen3 Next 80B A3B ThinkingAlibabaApache 2.0JSON+2 | $1.20/1Mvia OpenRouter | 262K | $0.15/1M | $1.20/1M | — | — | |
| Qwen3 VL 235B A22B ThinkingAlibabaApache 2.0JSON+3 | $3.95/1Mvia DevPass (LLM Gateway) | 131K | $0.98/1M | $3.95/1M | — | — | |
| Qwen3 VL 30B A3B InstructAlibabaApache 2.0JSON+3 | $0.389/1Mvia SiliconFlow | 262K | $0.097/1M | $0.389/1M | — | — | |
| Qwen3-VL PlusAlibabaPrompt cachingJSON+3 | $1.37/1Mvia AIHubMix | 262K | $0.143/1M | $1.43/1M | — | — | |
| Qwen3.5 122B-A10BAlibabaApache 2.0JSON+3 | $0.901/1Mvia AIHubMix | 262K | $0.40/1M | $3.20/1M | 1418 | — | |
| Qwen3.5 FlashAlibabaPrompt cachingJSON+3 | $0.26/1Mvia OpenRouter | 1M | $0.029/1M | $0.287/1M | 1397 | — | |
| Qwen3.5 PlusAlibabaJSONReasoning+2 | $0.658/1Mvia AIHubMix | 1M | $0.115/1M | $0.688/1M | — | — | |
| Qwen3.6 FlashAlibabaJSONReasoning+2 | $0.823/1Mvia APIMart | 1M | $0.165/1M | $0.99/1M | — | — | |
| Qwen3.6 Max PreviewAlibabaPrompt cachingJSON+2 | $9.98/1Mvia CometAPI | 256K | $1.66/1M | $9.98/1M | 1482 | — | |
| Qwen3.6 PlusAlibabaPrompt cachingJSON+3 | $1.37/1Mvia APIMart | 1M | $0.276/1M | $1.65/1M | 1462 | — | |
| Qwen3.7 FlashAlibabaPrompt cachingJSON+3 | $0.091/1Mvia APIMart | 1M | $0.028/1M | $0.11/1M | — | — | |
| Qwen3.8 Max PreviewAlibabaPrompt cachingJSON+3 | $6/1Mvia Charm Hyper | 262K | $2/1M | $6/1M | — | — | |
| Reka Flash 3Reka AIJSONReasoning | $0.20/1Mvia OpenRouter | 66K | $0.10/1M | $0.20/1M | — | — | |
 Ring 1tInclusionaiMITJSON+2 | $2.19/1Mvia AIHubMix | — | $0.548/1M | $2.19/1M | — | — | |
| Ring Flash 2.0InclusionaiMITJSON+2 | $0.544/1Mvia AIHubMix | — | $0.136/1M | $0.544/1M | — | — | |
| Ring-2.6-1TInclusionaiMITPrompt caching+3 | $1.10/1Mvia ZenMux | 262K | $0.132/1M | $1.10/1M | — | — | |
 Sakana NamazuSakana AIPrompt cachingJSON+3 | $4/1Mvia OpenRouter | 262K | $0.95/1M | $4/1M | — | — | |
| SSao10K Stheno 8bSao10KCC BY-NC 4.0Reasoning | $0.05/1Mvia Hugging Face | 8K | $0.05/1M | $0.05/1M | — | — | |
 Sarvam-105BSarvam AIReasoningTool calling | $0.16/1Mvia FastRouter | 128K | $0.04/1M | $0.16/1M | — | — | |
 Seed 1.6ByteDancePrompt cachingJSON+3 | $0.285/1Mvia APIYI | 262K | $0.114/1M | $0.285/1M | — | — | |
| Seed 1.6 FlashByteDanceJSONReasoning+2 | $0.30/1Mvia APIYI | 262K | $0.075/1M | $0.30/1M | — | — | |
| Seed 2.1 TurboByteDancePrompt cachingJSON+3 | $1.67/1Mvia CometAPI | 256K | $0.50/1M | $2.50/1M | — | — | |
| Seed EvolvingByteDancePrompt cachingJSON+3 | $3.33/1Mvia CometAPI | 256K | $0.667/1M | $3.33/1M | — | — | |
| Seed Oss 36B InstructByteDanceApache 2.0JSON+3 | $0.534/1Mvia AIHubMix | 256K | $0.20/1M | $0.534/1M | — | — | |
 Solar Pro3UpstagePrompt cachingJSON+2 | $0.60/1Mvia OpenRouter | 131K | $0.15/1M | $0.60/1M | — | — | |
| Sonar Pro SearchPerplexityJSONReasoning+1 | $15/1Mvia OpenRouter | 200K | $3/1M | $15/1M | — | — | |
| Sonar Reasoning ProPerplexityReasoningVision | $8/1Mvia FastRouter | 128K | $2/1M | $8/1M | — | — | |
 Step 3.5 FlashStepFunApache 2.0Reasoning+1 | $0.292/1Mvia SiliconFlow | 256K | $0.097/1M | $0.292/1M | — | — | |
| Step 3.7 FlashStepFunApache 2.0Prompt caching+4 | $0.925/1Mvia APIMart | 262K | $0.154/1M | $0.925/1M | — | — | |
| Step 5 PreviewStepFunOpen weightsPrompt caching+4 | $2.70/1Mvia Deep Infra | 1.05M | $1/1M | $2.70/1M | 1568 | — | |
 WizardLM-2 8x22BMicrosoftReasoning | $0.62/1Mvia Hugging Face | 66K | $0.62/1M | $0.62/1M | — | — |
Frequently asked questions
A model trained to work through a problem in intermediate steps, often called thinking, before it answers. It does better on math, coding and planning, at the cost of more tokens and slower replies.
The thinking steps are billed as output tokens, so one answer can use many times the tokens of a direct reply. Compare output prices, and check whether the model lets you limit or switch off its reasoning.
For short lookups, classification, extraction or chat, a regular model is faster and cheaper with little loss in quality. Keep reasoning for problems that really need several steps.
Can't find what you're looking for? Contact us