AI models: quality against real cost

169 models priced across every host serving them · 8 are undominated

Hover any point to identify the model. Click to keep it selected. Click again, press Escape, or click the background to release it.
6008001000120014001600$0.10$0.20$0.50$1$2$5$10$20gpt-oss-20b (batch)$0.062 · 896Ministral 3 8B 2512 (batc…$0.064 · 1039Solar Mini 4$0.088 · 1158GLM 5.3 Flash (batch)$0.097 · 1289MiMo-V2.6-Pro$0.472 · 1328GPT-6.1 Sol (batch)$2.02 · 1372Claude Sonnet 5.5 (batch)$2.02 · 1382Claude Opus 5.5 (batch)$4.03 · 1401Anthropic: Claude Haiku 5.5Anthropic: Claude Haiku 5.5 (batch)Mistral: Mistral Large 4OpenAI: GPT-6.1 SolOpenAI: GPT-6.1 Sol (batch)Anthropic: Claude Sonnet 5.5Anthropic: Claude Sonnet 5.5 (batch)Upstage: Solar Mini 4Anthropic: Claude Opus 5.5Anthropic: Claude Opus 5.5 (batch)Xiaomi: MiMo-V2.6-ProSpaceXAI: Grok 4.7Meta: Muse Spark 1.3Google: Gemini 3.8 FlashGoogle: Gemini 3.8 Flash (batch)Anthropic: Claude Fable 5.1Anthropic: Claude Fable 5.1 (batch)Z.ai: GLM 5.3 FlashZ.ai: GLM 5.3 Flash (batch)Z.ai: GLM 5.3Z.ai: GLM 5.3 (batch)Google: Gemini 3.7 FlashGoogle: Gemini 3.7 Flash (batch)SpaceXAI: Grok 4.6Upstage: Solar Pro 4Meta: Muse Spark 1.2DeepSeek: DeepSeek V4 Flash 0731Anthropic: Claude Opus 5Anthropic: Claude Opus 5 (batch)Google: Gemini 3.6 FlashGoogle: Gemini 3.6 Flash (batch)Thinking Machines: InklingMoonshotAI: Kimi K3MoonshotAI: Kimi K3 (batch)Meta: Muse Spark 1.1SpaceXAI: Grok 4.5Tencent: Hy3Anthropic: Claude Sonnet 5Anthropic: Claude Sonnet 5 (batch)Z.ai: GLM 5.2MoonshotAI: Kimi K2.7 CodeAnthropic: Claude Fable 5Anthropic: Claude Fable 5 (batch)NVIDIA: Nemotron 3 UltraQwen: Qwen3.7 PlusMiniMax: MiniMax M3StepFun: Step 3.7 FlashAnthropic: Claude Opus 4.8Anthropic: Claude Opus 4.8 (batch)Qwen: Qwen3.7 MaxGoogle: Gemini 3.5 FlashGoogle: Gemini 3.5 Flash (batch)SpaceXAI: Grok 4.3SpaceXAI: Grok 4.3 (batch)OpenAI: GPT-5.5OpenAI: GPT-5.5 (batch)DeepSeek: DeepSeek V4 Pro 0423DeepSeek: DeepSeek V4 Flash 0423Xiaomi: MiMo-V2.5-ProXiaomi: MiMo-V2.5MoonshotAI: Kimi K2.6Z.ai: GLM 5.1Qwen: Qwen3.6 PlusZ.ai: GLM 5V TurboArcee AI: Trinity Large ThinkingSpaceXAI: Grok 4.20MiniMax: MiniMax M2.7Z.ai: GLM 5 TurboOpenAI: GPT-5.4OpenAI: GPT-5.4 (batch)Inception: Mercury 2Google: Gemini 3.1 Flash Lite PreviewGoogle: Nano Banana 2 (Gemini 3.1 Flash Image Preview)OpenAI: GPT-5.3-CodexGoogle: Gemini 3.1 Pro PreviewGoogle: Gemini 3.1 Pro Preview (batch)Anthropic: Claude Sonnet 4.6Anthropic: Claude Sonnet 4.6 (batch)Qwen: Qwen3.5 Plus 2026-02-15Qwen: Qwen3.5 397B A17BMiniMax: MiniMax M2.5Z.ai: GLM 5Anthropic: Claude Opus 4.6Anthropic: Claude Opus 4.6 (batch)MoonshotAI: Kimi K2.5Z.ai: GLM 4.7 FlashMiniMax: MiniMax M2.1Z.ai: GLM 4.7Google: Gemini 3 Flash PreviewGoogle: Gemini 3 Flash Preview (batch)OpenAI: GPT-5.2OpenAI: GPT-5.2 (batch)Mistral: Ministral 3 14B 2512Mistral: Ministral 3 8B 2512Mistral: Ministral 3 8B 2512 (batch)Mistral: Ministral 3 3B 2512Mistral: Mistral Large 3 2512Mistral: Mistral Large 3 2512 (batch)DeepSeek: DeepSeek V3.2Anthropic: Claude Opus 4.5Anthropic: Claude Opus 4.5 (batch)Google: Nano Banana Pro (Gemini 3 Pro Image Preview)OpenAI: GPT-5.1OpenAI: GPT-5.1 (batch)OpenAI: GPT-5.1-CodexOpenAI: GPT-5.1-Codex-MiniMoonshotAI: Kimi K2 ThinkingAmazon: Nova Premier 1.0MiniMax: MiniMax M2OpenAI: GPT-5 Image MiniAnthropic: Claude Haiku 4.5Anthropic: Claude Haiku 4.5 (batch)OpenAI: GPT-5 ImageGoogle: Nano Banana (Gemini 2.5 Flash Image)Z.ai: GLM 4.6Anthropic: Claude Sonnet 4.5Anthropic: Claude Sonnet 4.5 (batch)DeepSeek: DeepSeek V3.2 ExpDeepSeek: DeepSeek V3.1 TerminusMoonshotAI: Kimi K2 0905DeepSeek: DeepSeek V3.1Mistral: Mistral Medium 3.1Mistral: Mistral Medium 3.1 (batch)OpenAI: GPT-5OpenAI: GPT-5 (batch)OpenAI: GPT-5 MiniOpenAI: GPT-5 Mini (batch)OpenAI: GPT-5 NanoOpenAI: GPT-5 Nano (batch)OpenAI: gpt-oss-120bOpenAI: gpt-oss-120b (batch)OpenAI: gpt-oss-20bOpenAI: gpt-oss-20b (batch)Anthropic: Claude Opus 4.1Anthropic: Claude Opus 4.1 (batch)Mistral: Codestral 2508Mistral: Codestral 2508 (batch)Qwen: Qwen3 Coder 30B A3B InstructZ.ai: GLM 4.5Z.ai: GLM 4.5 AirQwen: Qwen3 235B A22B Thinking 2507Qwen: Qwen3 Coder 480B A35BQwen: Qwen3 235B A22B Instruct 2507MoonshotAI: Kimi K2 0711Mistral: Mistral Small 3.2 24BGoogle: Gemini 2.5 FlashGoogle: Gemini 2.5 Flash (batch)Google: Gemini 2.5 ProGoogle: Gemini 2.5 Pro (batch)DeepSeek: R1 0528Anthropic: Claude Sonnet 4Mistral: Mistral Medium 3Qwen: Qwen3 30B A3BOpenAI: o3OpenAI: o3 (batch)OpenAI: o4 MiniOpenAI: o4 Mini (batch)OpenAI: GPT-4.1OpenAI: GPT-4.1 (batch)OpenAI: GPT-4.1 MiniOpenAI: GPT-4.1 Mini (batch)OpenAI: GPT-4.1 NanoOpenAI: GPT-4.1 Nano (batch)Meta: Llama 4 MaverickMeta: Llama 4 ScoutDeepSeek: DeepSeek V3Amazon: Nova Pro 1.0OpenAI: GPT-4oOpenAI: GPT-4o (batch)Quality: Arena Elo, higher is betterCost per 1,000 tasks, log scale, lower is better
8 best value, nothing is cheaper and as good161 beaten, by price:cheapest thirdmiddle thirddearest thirdEmpty space above the dashed line is what no model has reached yet.
Compiled 12 October 2026 · free to embedBestHunt AI model leaderboard →