The most intelligence for the money, right now
Every model below is one no other model beats on both counts: none scores higher for less. The blended price is a million tokens at three input to one output, the usual mix of a chat or an agent; cached input is what a repeated prompt prefix costs instead. Every figure is read from OpenRouter's model feed every 6 hours, so the table changes when a lab reprices.
| Model | Lab | Index | Blended / 1M | Input / 1M | Output / 1M | Cached input / 1M |
|---|---|---|---|---|---|---|
| MiMo-V2.6-Flash | Xiaomi | 26.7 | $0.175 | $0.14 | $0.28 | $0.0028 |
| Claude Haiku 5.5 | Anthropic | 36.2 | $0.20 | $0.10 | $0.50 | $0.01 |
| GPT-6.1 Terra | OpenAI | 51 | $2 | $1 | $5 | $0.05 |
| GPT-6.1 Sol | OpenAI | 54.5 | $4 | $2 | $10 | $0.10 |
| Claude Opus 5.5 | Anthropic | 58.4 | $8 | $4 | $20 | $0.20 |
Every guide
- Muse Code Pricing 2026: Plans, Real Limits, Wonder Full: Muse Code pricing 2026: plans from $5, real usage limits, and why turning off Contributor mode cuts your allowance about 12x. Pick your data mode first.
- Codex Pricing 2026: Plans, Limits, Resets and Credits: Compare every Codex plan, limit, reset and credit cost for 2026. Pick the right tier and stretch your allowance before October changes.
- Prompt Caching: Cut AI Agent Bills 45% in 2026: Reduce AI agent costs by 45% using prompt caching. Learn to reuse context in iterative loops and stop paying for redundant data processing.
- Grok Bot Pricing 2026: Real Costs, Limits, Alternatives: Compare Grok pricing from $8 to $300 tiers. Discover hidden limits and the best alternatives for xAI's always-on Grok Bot in 2026.
- GPT-6 Astra Pricing: Real Cost per Agent Task 2026: Master GPT-6 Astra costs with our bottom-up model for agent tasks, effort multipliers, and the 272K context cliff analysis.
- Gemini 3.8 Flash: $0.58/Task for Agents 2026: Gemini 3.8 Flash costs 40% more per task despite flat pricing. Learn why cost per token hides your real AI agent bill in 2026.
- Cheapest LLM APIs for AI Agents: 2026 Price Table: Save 90% on AI agent costs. Compare the cheapest 2026 LLM API prices, prompt caching, and routing strategies in our latest guide.
- Grok 4.6: Cheapest Frontier LLM for Agents 2026: Grok 4.6 delivers flagship reasoning at 60% less than GPT-5.6, setting a new price-performance standard for scaling AI agent fleets.
- Cut Your DeepSeek Agent Bill 50%: Off-Peak 2026: Master DeepSeek's time-of-day pricing. Learn to cut agent costs by 50% using smart scheduling, caching, and routing strategies.
- AI Model Routing in 2026: Cut Agent Costs 60%: Route each AI request to the cheapest capable model, stack caching and trimming, and slash agent costs by over 60% in 2026.
- How to Cut LLM Costs: The 2026 Efficiency Guide: Slash your LLM bill by 90% with proven 2026 strategies: API pricing, subsidized plans, caching, and model right-sizing techniques.
- Claude Code Pricing: What It Costs You (2026): Claude Code costs $13/active day on average. See how Max plans beat API billing by 90% for daily users in our full pricing breakdown.
- The True Cost of LLM Inference in 2026: AI token costs collapsed 1,200x in 5 years. Discover how companies lose billions subsidizing cheap intelligence and what it means for AI's future.
- Claude Agent SDK $200 Credit: The 2026 Guide: Discover what Anthropic's $200 monthly Agent SDK credit really costs and how to maximize every token in 2026.
- OpenClaw Pricing: What does it cost you (2026): Explore OpenClaw AI pricing for 2026: costs, setup options, and top alternatives to boost your productivity.
- ChatGPT Operator Pricing: What Does It Cost You (2026): Explore ChatGPT Operator pricing, 2026 plans, and automation costs to choose the best solution for your business needs.
- Claude Cowork Pricing 2026: Plans, Real Limits, Opus 5: Explore 2026 Claude Cowork pricing. Master Opus 5 shifts and Team plans starting at two seats. Get verified limits and cost details here.
- Claude Code Pricing 2026: Plans, Limits & Alternatives: Verified August 2026 Claude Code pricing. Compare subscription plans vs. API costs, usage limits, and model performance benchmarks.
- Appian Pricing Guide 2026: What Does It Cost You?: Appian's 2026 pricing remains opaque with high enterprise costs - explore AI-driven alternatives and hidden fees before committing
- AgentForce Pricing 2026: What Does It Cost You: Complete guide to AgentForce pricing models, hidden costs, and alternatives to help you make smart AI agent investment decisions
- Blue Prism Pricing: What Does It Cost You (2026): Blue Prism costs $10k-20k per bot annually, but AI-native automation platforms offer similar capabilities at 80% lower costs
- Browser Automation: What Does It Cost You? (Full 2026 Guide): Complete guide to browser automation costs in 2025: from free DIY tools to AI agents, helping you choose the right solution to save time and money
- Airtable AI Agents Pricing: what does it cost you (2026): Learn how Airtable AI Agents pricing works: 500 free credits per user, no extra fees, and when to upgrade for more AI automation power
- Automation Anywhere AI Agents Pricing: What Does It Cost You (2026): Comprehensive guide to Automation Anywhere AI Agents pricing, plans and alternatives to help you choose the right automation solution
- UiPath AI Agents Pricing: What It Really Costs You (2026): Master UiPath AI agent costs with 2026 metering rates. Learn model tier math to negotiate better enterprise contracts and optimize ROI.
- The Cost of AI Agents: Uncovering the True Cost of Agentic AI (2025 Report): Deep-dive into AI agents in 2025: pricing, platforms, use cases and deployment strategies to maximize ROI while avoiding costly pitfalls
- UiPath pricing: what does it cost you (2026): Uncover the real costs of UiPath automation in 2024, from hidden fees to ROI, and learn how to maximize your automation investment
Other topics
- Which AI model should your agents run on?
- Which AI coding agent should you use?
- How do AI agents use the browser and the computer?
- How do you build an AI agent?
- Which AI agents can do work for your business?
- How do you measure what AI can do?
- How do you keep AI agents safe?
- What is happening in AI, and what does it mean?