# Toknix Documentation (llms.txt) > Toknix is an AI API infrastructure platform. One API key reaches 34 foundation models from Anthropic, OpenAI, Google, xAI, DeepSeek, Zhipu AI (GLM), and Moonshot AI (Kimi) through OpenAI- and Anthropic-compatible endpoints at https://api.toknix.com. ## Core Gateway Details - Base URL (OpenAI Chat Completions): `https://api.toknix.com/v1` - Base URL (OpenAI Responses API): `https://api.toknix.com/v1` - Base URL (Anthropic Messages API): `https://api.toknix.com` - Authentication Header (OpenAI): `Authorization: Bearer tk_live_...` - Authentication Header (Anthropic): `x-api-key: tk_live_...` - Rate Limits: 120 requests/minute and 32 concurrent streams per account - Subkeys: Support for isolated sub-keys via `POST /v1/subkeys` - Billing: Prepaid USD balance, charged strictly per token with zero subscriptions. Top-ups from $5. - Payment: Crypto only (BTC, ETH, USDT, USDC, LTC, SOL). Instant automated deposits. - Documentation: https://docs.toknix.in ## Models & Pricing (Per 1M Tokens) ### Gemini - `gemini-4-carbon`: 1.05M Context | Input: $1.00 | Cached: $0.10 | Output: $1.00 - `gemini-4-argon`: 128K Context | Input: $0.60 | Cached: $0.10 | Output: $0.60 ### GPT - `gpt-6-astra`: 1.05M Context | Input: $1.30 | Cached: $0.12 | Output: $1.30 - `gpt-6.2-astra`: 1.05M Context | Input: $1.30 | Cached: $0.12 | Output: $1.30 - `gpt-6.1-sol`: 1.05M Context | Input: $0.30 | Cached: $0.08 | Output: $0.30 - `gpt-6.2-sol`: 1.05M Context | Input: $0.30 | Cached: $0.08 | Output: $0.30 - `gpt-6-sol`: 1.05M Context | Input: $0.30 | Cached: $0.08 | Output: $0.30 - `gpt-6-luna`: 1.05M Context | Input: $0.10 | Cached: $0.02 | Output: $0.10 - `gpt-5.6-sol`: 1M Context | Input: $0.90 | Cached: $0.08 | Output: $0.90 - `gpt-5.6-terra`: 1M Context | Input: $0.90 | Cached: $0.08 | Output: $0.90 - `gpt-5.6-luna`: 1M Context | Input: $0.90 | Cached: $0.08 | Output: $0.90 - `gpt-5.5`: 1M Context | Input: $0.90 | Cached: $0.08 | Output: $0.90 ### Claude - `claude-fable-5-2`: 1M Context | Input: $1.60 | Cached: $0.16 | Output: $1.60 - `claude-opus-5-5-max`: 1M Context | Input: $1.40 | Cached: $0.14 | Output: $1.40 - `claude-opus-5-5`: 1M Context | Input: $1.40 | Cached: $0.14 | Output: $1.40 - `claude-opus-5`: 1M Context | Input: $1.00 | Cached: $0.10 | Output: $1.00 - `claude-opus-4-8`: 1M Context | Input: $1.00 | Cached: $0.10 | Output: $1.00 - `claude-opus-4-7`: 1M Context | Input: $1.00 | Cached: $0.10 | Output: $1.00 - `claude-opus-4-6`: 1M Context | Input: $1.00 | Cached: $0.10 | Output: $1.00 - `claude-sonnet-5-5`: 1M Context | Input: $1.00 | Cached: $0.12 | Output: $1.00 - `claude-sonnet-5`: 1M Context | Input: $1.00 | Cached: $0.10 | Output: $1.00 - `claude-haiku-5-5`: 1M Context | Input: $0.06 | Cached: $0.02 | Output: $0.06 ### Grok - `grok-4.7`: 500K Context | Input: $0.20 | Cached: $0.04 | Output: $0.20 - `grok-4.6`: 500K Context | Input: $0.20 | Cached: $0.04 | Output: $0.20 - `grok-4.5`: 200K Context | Input: $0.20 | Cached: $0.04 | Output: $0.20 ### DeepSeek - `deepseek-v4.1-flash`: 1M Context | Input: $0.04 | Cached: $0.010 | Output: $0.04 - `deepseek-v4.1-flash-1017`: 1M Context | Input: $0.04 | Cached: $0.010 | Output: $0.04 - `deepseek-v4.1-flash-1017-hq`: 1M Context | Input: $0.10 | Cached: $0.02 | Output: $0.10 - `deepseek-v4.1-flash-1017-abliterated`: 1M Context | Input: $0.30 | Cached: $0.10 | Output: $0.30 - `deepseek-v4.1-flash-1017-abliterated-fast`: 1M Context | Input: $1.40 | Cached: $0.32 | Output: $1.40 - `deepseek-v4-flash-0731`: 1M Context | Input: $0.06 | Cached: $0.0054 | Output: $0.06 - `deepseek-v4-pro-0813`: 1M Context | Input: $0.50 | Cached: $0.10 | Output: $0.50 ### GLM - `glm-5.2`: 200K Context | Input: $0.12 | Cached: $0.034 | Output: $0.12 - `glm-5.3`: 200K Context | Input: $0.10 | Cached: $0.02 | Output: $0.10 - `glm-5.3-uncensored`: 200K Context | Input: $0.10 | Cached: $0.02 | Output: $0.10 - `glm-5.3-flash`: 1M Context | Input: $0.03 | Cached: $0.003 | Output: $0.03 - `glm-5.3-flash-uncensored`: 1M Context | Input: $0.05 | Cached: $0.005 | Output: $0.05 - `glm-5.3-abliterated`: 1M Context | Input: $1.60 | Cached: $0.50 | Output: $1.60 ### Kimi - `kimi-k3`: 1M Context | Input: $0.17 | Cached: $0.04 | Output: $0.17 - `kimi-k3.1-pro`: 1M Context | Input: $0.60 | Cached: $0.10 | Output: $0.60 ## Key Endpoints - `POST /v1/chat/completions` — OpenAI format chat completions with SSE streaming support - `POST /v1/responses` — OpenAI Responses API for modern agent harnesses - `POST /v1/messages` — Anthropic Messages format - `GET /v1/models` — Programmatic model catalog - `GET /v1/balance` — Current wallet balance check - `GET /v1/usage` — Token usage and prompt cache telemetry - `POST /v1/subkeys` — Create dedicated subkeys with custom balance allocations