Every Kodine model, one balance
Public per-token pricing, no hidden multipliers. Prices are per 1M tokens in USD.
| Model | Category | Best for | Input | Output | Context |
|---|---|---|---|---|---|
| Auto | Economy | Auto — routes to the best model for your plan | $0 / 1M | $0 / 1M | 1000K tokens |
| GPT-6 Astra | Premium | Hardest reasoning & long refactors | $6 / 1M | $30 / 1M | 872K tokens |
| Claude Opus 5.5 | Premium | Sharp analysis & polished writing | $4 / 1M | $20 / 1M | 1000K tokens |
| Claude Opus 5 | Premium | Previous-gen Opus power | $4 / 1M | $20 / 1M | 1000K tokens |
| Claude Opus 4.8 | Premium | Opus-class reasoning | $4 / 1M | $20 / 1M | 1000K tokens |
| Claude Opus 4.7 | Premium | Opus-class reasoning, great value | $3.50 / 1M | $17.50 / 1M | 1000K tokens |
| GPT-6.1 Sol | Premium | Everyday coding balance | $3 / 1M | $15 / 1M | 872K tokens |
| Claude Opus 4.6 | Premium | Solid Opus workhorse | $3 / 1M | $15 / 1M | 1000K tokens |
| Claude Opus 4.5 | Premium | Compact Opus, 200k context | $3 / 1M | $15 / 1M | 200K tokens |
| GPT-6 Sol | Premium | GPT-6 class for everyday work | $3 / 1M | $15 / 1M | 872K tokens |
| Claude Sonnet 5 | Premium | Fast frontier coding | $2.50 / 1M | $12 / 1M | 1000K tokens |
| Gemini Pro Agent | Premium | Google's agentic flagship | $2.50 / 1M | $12 / 1M | 1000K tokens |
| GPT-5.6 Sol | Standard | Code-heavy workflows | $2 / 1M | $10 / 1M | 1000K tokens |
| Claude Sonnet 4.6 | Standard | Reliable Sonnet coding | $2 / 1M | $10 / 1M | 1000K tokens |
| Gemini 3.1 Pro (Low) | Standard | Gemini Pro, low effort | $1.50 / 1M | $7 / 1M | 1000K tokens |
| Claude Sonnet 4.5 | Standard | Compact Sonnet | $1.50 / 1M | $7.50 / 1M | 200K tokens |
| DeepSeek V4 Pro | Standard | DeepSeek pro reasoning | $1 / 1M | $4 / 1M | 256K tokens |
| Qwen 3.8 Max 0902 | Standard | Qwen Max, pinned 0902 snapshot | $1 / 1M | $4 / 1M | 256K tokens |
| Qwen 3.8 Max | Standard | Strong open-weight coding | $1 / 1M | $4 / 1M | 256K tokens |
| Gemini 3.7 Flash (High) | Standard | High-effort fast Gemini | $0.80 / 1M | $3 / 1M | 1000K tokens |
| GPT-5.6 Terra | Standard | Fast interactive loops | $1 / 1M | $4 / 1M | 1000K tokens |
| Gemini 3.6 Flash (High) | Standard | Fast Gemini, high effort | $0.60 / 1M | $2.50 / 1M | 1000K tokens |
| GPT-6 Luna | Economy | Cheapest GPT-6 | $1 / 1M | $3 / 1M | 872K tokens |
| GPT-5.6 Luna | Economy | Cheapest GPT-5.6 | $0.80 / 1M | $2.50 / 1M | 1000K tokens |
| Gemini 3 Flash | Economy | Light tasks at low cost | $0.30 / 1M | $1.20 / 1M | 1000K tokens |
| Gemini 3.1 Flash Lite | Economy | Ultra-cheap Gemini | $0.15 / 1M | $0.60 / 1M | 1000K tokens |
| Qwen 3.8 Flash | Economy | Most economical option | $0.20 / 1M | $0.50 / 1M | 256K tokens |
| DeepSeek V4.1 Flash | Economy | Fast DeepSeek | $0.30 / 1M | $1 / 1M | 256K tokens |
| DeepSeek v3.2 | Economy | DeepSeek v3.2 | $0.50 / 1M | $2 / 1M | 164K tokens |
| Kimi K3 | Economy | Moonshot's Kimi K3 | $0.60 / 1M | $2.50 / 1M | 256K tokens |
| GLM-5.3 | Economy | Zhipu GLM-5.3 | $0.60 / 1M | $2.50 / 1M | 256K tokens |
| Claude Haiku 4.5 | Economy | Fast, cheap Claude | $0.80 / 1M | $4 / 1M | 200K tokens |
Aura models are continuously benchmarked. The lineup changes as better models ship - prices only ever move down.
Hear about new releases first
Join the list and get early access.