Claude Code with any model
Claude Code talks to the Anthropic Messages API, and every model in our catalog answers on it. Set two variables and Claude Code runs on your Inferray key — on Claude, or on Kimi, GLM, DeepSeek or GPT — with no 5-hour or weekly windows.
Two variables
export ANTHROPIC_BASE_URL="https://api.inferray.com"
export ANTHROPIC_AUTH_TOKEN="inferray_..."
export ANTHROPIC_MODEL="claude-sonnet-5-5"
That is the whole change. Claude Code keeps working the way it does; it bills to your key instead of your Claude plan. The setup guide covers the settings file, the VS Code extension, and how to check it worked.
Pick a model
| Model | Input / 1M | Cached / 1M | Output / 1M | Context | Token Plans |
|---|---|---|---|---|---|
| Claude Opus 5.5 | $4 | $0.2 | $20 | 1M | Included |
| Claude Sonnet 5.5 | $2 | $0.1 | $10 | 1M | Included |
| Claude Haiku 5.5 | $0.1 | $0.01 | $0.5 | 1M | Included |
| Kimi K3 | $3 | $0.3 | $15 | 1.0M | Included |
| GLM-5.3 | $0.77 | $0.143 | $2.42 | 262K | Included |
| DeepSeek V4.1 Flash | $0.15 | $0.003 | $0.6 | 1M | Included |
| GPT-6.1 Sol | $2 | $0.1 | $10 | 1.1M | Included |
Change ANTHROPIC_MODEL to switch. A common split is a frontier model for planning and hard bugs and a lower-priced one for the bulk of the edits.
Pay for it with a Token Plan
| Plan | You pay / month | Usage at list price | ≈ tokens on Claude Sonnet 5.5 | ≈ tokens on Kimi K3 | ≈ tokens on GLM-5.3 |
|---|---|---|---|---|---|
| Solo | $20 | $60 | 96M | 56M | 210M |
| Pro | $100 | $300 | 480M | 280M | 1B |
| Max | $200 | $600 | 960M | 560M | 2.1B |
Token estimates assume coding-agent traffic: 85% cached input, 12% fresh input, 3% output.
No 5-hour session limit, no weekly cap: the month's credit is there from day one, and every request shows what it cost.
Claude Code with Kimi
Kimi K3 is Moonshot AI's model for long-horizon coding, with a 1.0M context. Set ANTHROPIC_MODEL="kimi-k3". It costs $3 per 1M input tokens and $15 per 1M output tokens; $60 of plan credit is about 56M tokens of agent traffic. More on the Kimi K3 page.
Claude Code with GLM
GLM-5.3 is Z.ai's coding-first flagship, with a 262K context. Set ANTHROPIC_MODEL="glm-5.3". It costs $0.77 in and $2.42 out per 1M tokens — about 210M tokens of agent traffic for $60. Compare it with Z.ai's own GLM Coding Plan.
Claude Code with DeepSeek
DeepSeek V4.1 Flash is the lowest-priced model here: $0.15 in and $0.6 out per 1M tokens. Set ANTHROPIC_MODEL="deepseek-v4.1-flash" for high-volume edits and refactors. See the DeepSeek API page.
Claude Code with GPT
GPT-6.1 Sol runs in Claude Code the same way: ANTHROPIC_MODEL="gpt-6.1-sol", at $2 in and $10 out per 1M tokens. If you would rather use OpenAI's own agent, see Codex with any model.
Prices on this page come from the model catalog and were current on November 12, 2026.