Skip to main content

AI models

158 modelsCompare API prices
/
GLM 5.3 FlashxPrompt cachingJSON+3$0.987/1Mvia AIHubMix1.05M$0.282/1M$0.987/1M7KSep 18
Kimi K2.8 PreviewPrompt cachingReasoning+1$3.91/1Mvia CrazyRouter1.05M$0.942/1M$3.91/1M5KSep 14
MiMo-V2.5-ProMITPrompt caching+3$0.609/1Mvia OpenRouter1.05M$0.304/1M$0.609/1M5KApr 27
GLM 5.3 PrimePrompt cachingJSON+2$8.23/1Mvia AIHubMix1M$2.35/1M$8.23/1M5KSep 23
Nemotron 3.5 Lightning 30B A3BReasoningTool calling$0.16/1Mvia Deep Infra1MFree/1MFree/1M5KAug 11
Qwen3.7 FlashJSON$0.11/1Mvia EmpirioLabs AI1M$0.028/1M$0.11/1M5K—
DeepSeek V4 Flash 0731MITPrompt caching+3$0.133/1Mvia OpenRouter1.05M$0.044/1M$0.133/1M4KJul 31
LongCat-2.0MITPrompt caching+2$1.20/1Mvia Atlas Cloud1.05M$0.30/1M$1.20/1M4KJul 5
Qwen3.7 MaxPrompt cachingJSON+2$3.75/1Mvia DevPass (LLM Gateway)1M$1.65/1M$4.95/1M3KMay 21
Qwen3.8 Max 0902Prompt cachingJSON+3$4.11/1Mvia APIMart1M$2/1M$6/1M3KSep 1
Qwen3.7 PlusPrompt cachingJSON+3$0.30/1Mvia CrazyRouter1M$0.276/1M$1.10/1M2KJun 2
Qwen 3.8 Max PrimePrompt cachingJSON+3$12/1Mvia OpenRouter1M$4/1M$12/1M2KSep 23
Nemotron 3 Ultra 550B A55BReasoningTool calling$2.20/1Mvia Deep Infra1MFree/1MFree/1M1KJun 4
Fugu MaxPrompt cachingJSON+3$6/1Mvia EmpirioLabs AI1M$2/1M$6/1M1KSep 10
Beam 1.0 501B A23BPrompt cachingJSON+2$1/1Mvia OpenRouter1.05M$0.25/1M$1/1M—Oct 16
Claude Opus 4.6Prompt cachingJSON+3$20/1Mvia APIMart1M$5/1M$25/1M—Feb 5
Claude Opus 4.7Prompt cachingJSON+3$20/1Mvia APIMart1M$5/1M$25/1M—Apr 16
Claude Opus 4.8Prompt cachingJSON+3$20/1Mvia APIMart1M$5/1M$25/1M—May 28
Claude Sonnet 4Prompt cachingJSON+3$15/1Mvia Atlas Cloud1M$3/1M$15/1M—May 22, 2025
Claude Sonnet 4.5 (latest)Prompt cachingJSON+3$12/1Mvia APIMart1M$3/1M$15/1M—Sep 29, 2025
DeepSeek ReasonerPrompt cachingJSON+2$0.28/1Mvia Requesty1.05M$0.14/1M$0.28/1M—Jan 20, 2025
DeepSeek V4 Flash 0423 (Thinking)Prompt cachingJSON+1$0.51/1Mvia Ofox1M$0.19/1M$0.51/1M—Apr 23
DeepSeek V4 Flash 0424ReasoningTool calling$0.20/1Mvia Requesty1.05M$0.10/1M$0.20/1M—Apr 24
DeepSeek V4 Flash 0731Prompt cachingJSON$1.32/1Mvia EmpirioLabs AI1M$0.44/1M$1.32/1M——
DeepSeek V4 Flash Vision ExpMITPrompt caching+4$0.48/1Mvia CometAPI1.05M$0.12/1M$0.48/1M—Aug 31
DeepSeek V4 ProMITPrompt caching+3$1.58/1Mvia CometAPI1M$0.66/1M$1.98/1M—Apr 22
DeepSeek V4 ProPrompt cachingJSON$4.80/1Mvia EmpirioLabs AI1M$2.40/1M$4.80/1M——
DeepSeek V4 ProPrompt cachingJSON$3.30/1Mvia EmpirioLabs AI1M$1.65/1M$3.30/1M——
DeepSeek V4 ProPrompt cachingJSON$4.80/1Mvia EmpirioLabs AI1M$2.40/1M$4.80/1M——
DeepSeek V4 Pro 0423Prompt cachingJSON+2$2.60/1Mvia Requesty1.05M$1.30/1M$2.60/1M—Apr 23
Ember-1Prompt cachingJSON+3$15/1Mvia Fireworks AI1.05M$3/1M$15/1M—Sep 23
Fugu UltraPrompt cachingJSON+3$30/1Mvia EmpirioLabs AI1M$5/1M$30/1M—Jun 21
Fugu Ultra v1.1Prompt cachingVision$30.30/1Mvia Poe1M$5.05/1M$30.30/1M——
Fugu Ultra v2.0Prompt cachingJSON+3$30/1Mvia EmpirioLabs AI1M$5/1M$30/1M—Sep 10
Gemini 2.5 Flash Lite Preview 09-2025Prompt cachingJSON+2$0.40/1Mvia AIHubMix1.05M$0.10/1M$0.40/1M——
Gemini 2.5 Flash Preview 0520Prompt cachingJSON+2$2.50/1Mvia AIHubMix1.05M$0.30/1M$2.50/1M——
Gemini 2.5 Flash-LitePrompt cachingJSON+3$0.32/1Mvia APIMart1.05M—$0.40/1M—Jun 17, 2025
Gemini 2.5 Pro Preview 0605Prompt cachingJSON+3$3/1Mvia Kie.ai1.05M$0.38/1M$3/1M—Jun 5, 2025
Gemini 3 Flash PreviewPrompt cachingJSON+3$1.50/1Mvia Vertex1.05M—$3/1M—Dec 17, 2025
Gemini 3 Pro PreviewPrompt cachingJSON+3$9.60/1Mvia APIMart1.05M$1.60/1M$9.60/1M—Feb 19
Gemini 3.1 Flash LitePrompt cachingJSON+3$1.20/1Mvia CometAPI1.05M—$1.50/1M—May 7
Gemini 3.1 Flash Lite PreviewPrompt cachingJSON+3$1.20/1Mvia APIMart1.05M$0.20/1M$1.20/1M—Mar 3
Gemini 3.1 Pro Preview Custom ToolsPrompt cachingJSON+3$10.80/1Mvia APIYI1.05M$1.80/1M$10.80/1M—Feb 25
Gemini 3.5 FlashPrompt cachingJSON+3$2.70/1Mvia Kie.ai1.05M$0.45/1M$2.70/1M—May 19
Gemini 3.5 Flash LitePrompt cachingJSON+3$2/1Mvia APIMart1.05M—$2.50/1M—Jul 21
Gemini 3.6 FlashPrompt cachingJSON+3$3/1Mvia CometAPI1.05M$0.75/1M$3.75/1M—Jul 21
Gemini 3.7 FlashPrompt cachingJSON+3$1.13/1Mvia Kie.ai1.05M$0.225/1M$1.13/1M—Aug 13
Gemini Pro LatestPrompt cachingJSON+3$6.90/1Mvia Kenari1.05M$2/1M$12/1M—Feb 19
GLM-5.2Prompt cachingJSON$3.85/1Mvia EmpirioLabs AI1M$1.10/1M$3.85/1M——
GLM-5.2Prompt cachingJSON$4.40/1Mvia EmpirioLabs AI1M$1.40/1M$4.40/1M——

Best AI models by category

The top three in each category by LMArena blind-vote score, with the lowest API price. Updated daily.

Best LLMs

Gemini Pro LatestGoogleLMArena1480Lowest price$6.90/1M
  1. 2Gemini 3 Pro1479
  2. 3Qwen3.7 Max Preview1475

Best for coding

Claude Opus 5.5AnthropicLMArena1807Lowest price$16/1M
  1. 2GPT-6 Astra1775
  2. 3Claude Sonnet 5.51767

Best image models

Qwen Image 3.0 ProAlibabaLMArena1255Lowest price$0.003/img
  1. 2Ideogram V41205
  2. 3Recraft V4.1 Pro1169

Best video models

Gemini-Omni-1.1-FlashGoogleLMArena1516Lowest price$0.10/run
  1. 2Gemini Omni Flash Preview1513
  2. 3Flux 31493

Best open-source LLMs

GLM-5Zhipu AILMArena1446Lowest price$1.28/1M
  1. 2Kimi K2.51445
  2. 3GLM-4.61440

AI model comparison: quality vs price

Every ranked LLM plotted by its LMArena score against its lowest price per 1M tokens. Models on the line give the most quality for the money.

Arena score vs. priceLowest output price per 1M tokens, log scale · up and to the left is better valueBest score at its priceOther models
100011001200130014001500$0.03$0.10$0.30$1$3$10$30$100Output price per 1M tokensQwen3.5 FlashQwen3.7 PlusGemini 3 FlashGemini Pro LatestGemma 2 9B ItGemma 3 12B ITGemma 3 4B ITGPT OSS 120B

Scores: LMArena Text leaderboard (CC BY 4.0). Prices: AIModelFind.

Best score at its priceNo cheaper model scores higher · score · lowest price per 1M
  1. Gemma 2 9B It1207$0.03
  2. Gemma 3 4B IT1291$0.05
  3. Gemma 3 12B IT1334$0.10
  4. GPT OSS 120B1366$0.14
  5. Qwen3.5 Flash1397$0.26
  6. Qwen3.7 Plus1455$0.30
  7. Gemini 3 Flash1466$2.40
  8. ERNIE 5.11468$2.54
  9. Gemini Pro Latest1480$6.90

Scores: LMArena Text leaderboard (CC BY 4.0). Prices: AIModelFind.

Browse AI models by type and use

Ready-made lists of free, open-source, Chinese and new models, and models for each task.

By model type

By task

Free, open and new