AI CTRL Ratecard - NextGen
- Updated on Oct 23, 2026
- Published on Oct 15, 2026
Universal Credits powered by GPT-6 Luna
Based on standard usage patterns with a 75:25 input-to-output ratio. Credits work across all models — token allocations vary by model capability.
OpenAIAnthropicGooglePerplexity⬆ More tokens than GPT-6 Luna
O
OpenAI
GPT-6 and GPT-5 series models
| Model | Input Tokens / 1000 CR | Output Tokens / 1000 CR |
|---|
GPT-6 LunaStandard
Fast, cost-efficient model for focused, high-volume tasks | 75M | 15M |
GPT Image 2
Latest image generation with rich detail, edits, and complex scenes | 1.5M | 250K |
GPT Image 2.5 Flare
Image generation | 1.5M | 250K |
GPT-4.1
Strong coding and long-context instruction following | 3.75M | 937.5K |
GPT-4.1 mini
Cost-effective coding, extraction, and everyday assistants | 18.75M | 4.69M |
GPT-4.1 nano
High-volume classification, routing, and low-latency tasks | 75M | 18.75M |
GPT-4o
Vision and general multimodal assistants | 3M | 750K |
GPT-4o mini
Budget multimodal chat, vision, and simple extraction | 50M | 12.5M |
GPT-5
Reasoning, coding, and general-purpose assistants | 6M | 750K |
GPT-5 mini
Legacy mid-tier model | 30M | 3.75M |
GPT-5 nano
Ultra-efficient, high-volume tasks | 150M | 18.75M |
GPT-5 Pro
Deep analysis and planning with extended thinking | 500K | 62.5K |
GPT-5.1
Everyday coding and agent-style tasks | 6M | 750K |
GPT-5.2 Pro
Very demanding reasoning, long-form analysis, complex decisions | 357.14K | 44.64K |
GPT-5.3 Codex
Agentic coding, refactors, and multi-file software work | 4.29M | 535.71K |
GPT-5.4
Coding, agents, math, and broad analysis | 3M | 500K |
GPT-5.4 Mini
Quick iterations, lighter coding, and tool-heavy workflows | 10M | 1.67M |
GPT-5.4 Nano
Snippets, small fixes, classification, and routing | 37.5M | 6M |
GPT-5.4 Pro
Hardest problems, maximum precision, long-running work | 250K | 41.67K |
GPT-5.5
Flagship for coding, agents, computer-use, and professional work | 1.5M | 250K |
GPT-5.5 Pro
Highest-precision reasoning, audits, and high-stakes professional work | 250K | 41.67K |
GPT-5.6 Luna
High-volume extraction, classification, drafts, and cost-sensitive tasks | 37.5M | 6.25M |
GPT-5.6 Terra
Balanced production coding, analysis, and everyday professional work | 3.75M | 625K |
GPT-5.6-Sol
Frontier flagship (Sol) for hardest reasoning, agents, and professional work | 1.88M | 375K |
GPT-6 Astra
Flagship reasoning model | 750K | 150K |
GPT-6 Sol
Complex coding and agentic workflows | 3.75M | 750K |
GPT-6.1 Sol
Complex coding, computer use, and professional work | 3.75M | 750K |
A
Anthropic
Claude Fable, Opus, Sonnet, and Haiku model families
| Model | Input Tokens / 1000 CR | Output Tokens / 1000 CR |
|---|
Claude Fable 5
Frontier Mythos-class reasoning with stronger safeguards for high-risk domains | 750K | 150K |
Claude Fable 5.1
Frontier reasoning, hardest tasks | 750K | 150K |
Claude Haiku 4.5
Fast, efficient production tasks | 7.5M | 1.5M |
Claude Opus 4.5
Elite reasoning, research depth, and strategic analysis | 1.5M | 300K |
Claude Opus 4.6
Hardest agents, complex coding, and very long documents | 1.5M | 300K |
Claude Opus 4.7
Stronger coding, vision, and complex multi-step tasks with improved consistency | 1.5M | 300K |
Claude Opus 4.8
Most capable Opus for complex reasoning, long-horizon agentic coding, and high-autonomy work | 1.5M | 300K |
Claude Opus 5
Deep reasoning, agentic coding, and long-horizon enterprise work | 1.5M | 300K |
Claude Opus 5.5
Most capable model in the 5.5 family | 1.88M | 375K |
Claude Sonnet 4.6
Coding, agent planning, and drafting with strong long-context reasoning | 2.5M | 500K |
Claude Sonnet 5
Legacy mid-tier model | 3.75M | 750K |
Claude Sonnet 5.5
Mid-tier model balancing speed and capability | 3.75M | 750K |
G
Google Gemini
Gemini Pro, Flash, and Flash-Lite models
| Model | Input Tokens / 1000 CR | Output Tokens / 1000 CR |
|---|
Gemini 2.5 Flash
Chat, summarization, and extraction at large scale | 25M | 3M |
Gemini 2.5 Flash-Lite
Routing, translation, and extremely high request volume | 75M | 18.75M |
Gemini 2.5 Pro
Complex code, deep analysis, and very long context | 6M | 750K |
Gemini 3 Flash
Fast multimodal work, thinking-style tasks, large context | 15M | 2.5M |
Gemini 3.1 Flash-Lite
High-volume, budget tier | 30M | 5M |
Gemini 3.1 Pro
Flagship reasoning, 2M context | 3.75M | 625K |
Gemini 3.5 Flash
Fast reasoning, multimodal agents, and large-scale production workloads | 5M | 833.33K |
Gemini 3.7 Flash
Workhorse Flash for coding, agents, and production workloads | 10M | 2M |
Gemini 3.8 Flash
Balanced speed and quality | 10M | 2M |
| Model | Input Tokens / 1000 CR | Output Tokens / 1000 CR |
|---|
Grok 4.7
Frontier model for coding, agentic tasks, and knowledge work | 3.75M | 1.25M |
P
Perplexity
Sonar search-grounded models
| Model | Input Tokens / 1000 CR | Output Tokens / 1000 CR |
|---|
Sonar
Fast web-grounded answers | 7.5M | 7.5M |
Sonar Deep Research
Long reports, due diligence, and literature-style synthesis | 3.75M | 937.5K |
Sonar Pro
Advanced search-augmented answers | 2.5M | 500K |
Sonar Reasoning Pro
Search + multi-step reasoning | 3.75M | 937.5K |
Understanding This Rate Card
- Higher-capability models consume more credits per token
- Efficient models stretch your credits further
- All values based on 1000 Credits
- GPT-6 Luna is the baseline standard model
- Green values indicate more tokens than GPT-6 Luna
- Rates are subject to change by providers
- Actual usage varies based on conversation patterns
© 2026 Expedient. All rights reserved. | AI CTRL Platform Documentation
Page last updated: 10/23/2026, 6:00:04 AM ET
Was this article helpful?