Leaderboard
The AI model leaderboard
Filter by use case, sort by performance, price or value. Tier = AIRaceBoard editorial capability tier (5 = frontier, 1 = basic), based on each lab's positioning and our review. Not a benchmark score.
| # | Model | Region | Tier | Price in / out (per 1M) | Context | Best for | |
|---|---|---|---|---|---|---|---|
| 1 | GPT-6 AstraOpenAI | US | Frontier | $10 / $50 | Unknown | Hard reasoning, research, high-stakes analysis | src |
| 2 | Claude Fable 5.1Anthropic | US | Frontier | $10 / $50 | 1M | Complex agents, long documents, deep analysis | src |
| 3 | Claude Opus 5.5Anthropic | US | Frontier | $4 / $20 | 1M | Vibe coding, refactors, long-form writing | src |
| 4 | Gemini 4 ArgonGoogle | US | Frontier | $2 / $10 | 1M | Hard reasoning, coding, long agentic work | src |
| 5 | Kimi K3Moonshot AI | CN | Frontier | $3 / $15 | 1M | Agents, coding, long context | src |
| 6 | GPT-6.1 SolOpenAI | US | Pro | $2 / $10 | Unknown | General assistant, coding, apps | src |
| 7 | Claude Sonnet 5.5Anthropic | US | Pro | $2 / $10 | 1M | Everyday coding, writing, agents | src |
| 8 | Grok 4.7xAI | US | Pro | $2 / $6 | 500K | Reasoning, real-time-aware apps, coding | src |
| 9 | GPT-5.3-CodexOpenAI | US | Pro | $1.75 / $14 | Unknown | Code generation & agents | src |
| 10 | DeepSeek V4 ProDeepSeek | CN | Pro | $1.32 / $3.96 | 1M | Budget reasoning & coding | src |
| 11 | Qwen3.8-MaxAlibaba (Qwen) | CN | Pro | $2 / $6 | 1M | General reasoning, multilingual | src |
| 12 | GLM-5.3Zhipu / Z.ai | CN | Pro | $1.4 / $4.4 | 1M | Coding agents, tool use | src |
| 13 | Grok 4.3xAI | US | Strong | $1.25 / $2.5 | 1M | High-volume chat, summarization | src |
| 14 | Gemini 3.8 FlashGoogle | US | Strong | $0.75 / $3.75 | 1M | Fast apps, multimodal | src |
| 15 | DeepSeek V4.1 FlashDeepSeek | CN | Strong | $0.3 / $1.2 | 1M | Bulk processing, bots | src |
| 16 | gpt-oss-120BOpenAI (via Together) | US | Strong | $0.15 / $0.6 | 131K | Self-host, private deployments | src |
| 17 | Claude Haiku 5.5Anthropic | US | Strong | $0.1 / $0.5 | 1M | High-volume support, routing | src |
| 18 | GPT-6 LunaOpenAI | US | Strong | $0.1 / $0.5 | 1M | Bulk tasks, bots, classification | src |
| 19 | MiniMax-M3MiniMax | CN | Strong | $0.3 / $1.2 | 1M | Bulk agents, long context on a budget | src |
| 20 | Gemini 3.5 Flash-LiteGoogle | US | Lite | $0.3 / $2.5 | 1M | Classification, extraction | src |
| 21 | Mistral Small 4Mistral | EU | Lite | $0.15 / $0.6 | 262K | Lightweight tasks, self-hosting | src |
| 22 | Qwen3.8 FlashAlibaba (Qwen) | CN | Lite | $0.15 / $0.47 | 1M | Classification, chat at scale | src |
| 23 | GLM-5.3-FlashZhipu / Z.ai | CN | Lite | $0.15 / $0.5 | 1M | High-volume simple tasks | src |