Leaderboards
Sorted by public fields, criteria stated, raw data verifiable. Ranking, not recommending.
Only models with prices, context between 8K and 10.5M (dirty metadata filtered).
Cheapest input
- 1Ambient SmallAmbient$0.001
- 2qwen3.8-27b-4qAmbient$0.001
- 3GPT OSS 20BKilo Gateway$0.004
- 4GPT OSS 20BNanoGPT$0.004
- 5DeepSeek: DeepSeek Flash LatestKilo Gateway$0.006
- 6DeepSeek Flash LatestOpenRouter$0.006
- 7GLM-5.2OpenRouter$0.006
- 8GPT OSS 20B (Greenference)Eden AI$0.009
- 9Google Gemma 2Helicone$0.01
- 10Llama 3.2 1B InstructInference$0.01
- 11inclusionAI: Ling 3.0 TinyKilo Gateway$0.015
- 12Ling 3.0 TinyNanoGPT$0.015
- 13Ling 3.0 Tiny ThinkingNanoGPT$0.015
- 14nemotron-3-superOllama Cloud$0.015
- 15Ling 3.0 TinyOpenRouter$0.015
- 16GPT OSS 120BOpenRouter$0.015
- 17Granite 4.0 H MicroCloudflare Workers AI$0.017
- 18GPT OSS 120BKilo Gateway$0.019
- 19Llama 3.1 8B InstructAbacus$0.02
- 20Gemma 4 E4B ITDeep Infra$0.02
Cheapest output
- 1Ambient SmallAmbient$0.001
- 2qwen3.8-27b-4qAmbient$0.001
- 3Llama 3.2 1B InstructInference$0.01
- 4GPT OSS 20BKilo Gateway$0.017
- 5GPT OSS 20BNanoGPT$0.017
- 6Llama 3.2 3B Instruct (Deep Infra)Eden AI$0.02
- 7Llama 3.2 3B InstructInference$0.02
- 8Llama 3.2 1B InstructNovita AI$0.02
- 9Llama 3.1 8B InstructInference$0.025
- 10Google Gemma 2Helicone$0.03
- 11Meta Llama 3.1 8B Instruct TurboHelicone$0.03
- 12Llama 3.1 8B (decentralized)NanoGPT$0.03
- 13Mistral NemoOpenRouter$0.03
- 14Mistral NemoVercel AI Gateway$0.03
- 15Ministral 3BAzure$0.04
- 16Ministral 3BAzure Cognitive Services$0.04
- 17Llama-3.1-8B-InstructKilo Gateway$0.04
- 18Mistral Nemo Instruct 2407Meganova$0.04
- 19Llama 3.1 8b InstructNanoGPT$0.04
- 20GPT OSS 20B (Greenference)Eden AI$0.045
Largest context
- 1Qwen LongAlibaba (China)10,000K
- 2Llama 4 Scout 17B InstructAmazon Bedrock10,000K
- 3Llama 4 Scout 17B Instruct (US)Amazon Bedrock10,000K
- 4Liquid AutoLiquid Inference10,000K
- 5DeepSeek V4.1 Flash MarathonSubconscious5,000K
- 6GLM-5.3 MarathonSubconscious5,000K
- 7Gemini 2.0 Pro 0205NanoGPT2,097K
- 8Gemini 2.0 Pro 1206NanoGPT2,097K
- 9gemini-2.0-flash-lite302.AI2,000K
- 10grok-4-1-fast-reasoning302.AI2,000K
- 11grok-4-fast-non-reasoning302.AI2,000K
- 12grok-4-fast-reasoning302.AI2,000K
- 13grok-4.20-beta-0309-reasoning302.AI2,000K
- 14Grok 4.1 Fast (Non-Reasoning)Abacus2,000K
- 15Grok 4 Fast (Non-Reasoning)Abacus2,000K
- 16Grok 4.20 Non-ReasoningAI/ML API2,000K
- 17Grok 4 Fast Non-ReasoningAI/ML API2,000K
- 18Grok 4.20 (Non-Reasoning)Cloudflare AI Gateway2,000K
- 19Grok 4.20 (Reasoning)Cloudflare AI Gateway2,000K
- 20X-Ai/Grok-4-Fast-Non-ReasoningEden AI2,000K
Largest max output
- 1X-Ai/Grok-4-Fast-Non-ReasoningEden AI2,000K
- 2xAI Grok 4.1 Fast ReasoningHelicone2,000K
- 3xAI Grok 4 Fast Non-ReasoningHelicone2,000K
- 4xAI: Grok 4 Fast ReasoningHelicone2,000K
- 5grok-4-fast-non-reasoningJiekou.AI2,000K
- 6grok-4-fast-reasoningJiekou.AI2,000K
- 7Grok 4.20 Multi-AgentVercel AI Gateway2,000K
- 8Grok 4.20 Multi Agent BetaVercel AI Gateway2,000K
- 9Grok 4.20 Non-ReasoningVercel AI Gateway2,000K
- 10Grok 4.20 Beta Non-ReasoningVercel AI Gateway2,000K
- 11Grok 4.20 ReasoningVercel AI Gateway2,000K
- 12Grok 4.20 Beta ReasoningVercel AI Gateway2,000K
- 13Auto RouterOpenRouter2,000K
- 14x-AI/Grok-4-FastQiniu2,000K
- 15X-Ai/Grok-4-Fast-Non-ReasoningQiniu2,000K
- 16X-Ai/Grok-4-Fast-ReasoningQiniu2,000K
- 17x-AI/Grok-4.1-FastQiniu2,000K
- 18X-Ai/Grok 4.1 Fast Non ReasoningQiniu2,000K
- 19X-Ai/Grok 4.1 Fast ReasoningQiniu2,000K
- 20Grok 4.20 Non-ReasoningAI/ML API1,999K