# Tera - [Introduction](https://docs.tera.gw/introduction.md): Tera is an OpenAI-compatible inference API serving high-quality open-weight models. - [Quickstart](https://docs.tera.gw/quickstart.md): Make your first inference request in under a minute. - [Authentication](https://docs.tera.gw/authentication.md): How Tera API keys work. - [Privacy](https://docs.tera.gw/privacy.md): Zero retention. Zero training. Zero human review. - [OpenAI compatibility](https://docs.tera.gw/concepts/openai-compat.md): What carries over from OpenAI clients, and what's different. - [Streaming](https://docs.tera.gw/concepts/streaming.md): Token-by-token output over Server-Sent Events. - [Reasoning models](https://docs.tera.gw/concepts/reasoning.md): How to use chain-of-thought traces with thinking models on Tera. - [Tool calling](https://docs.tera.gw/concepts/tool-calling.md): OpenAI-compatible function calling. - [Claude Code](https://docs.tera.gw/guides/claude-code.md): Run Claude Code on any Tera model. - [Pricing](https://docs.tera.gw/pricing.md): Per-token pricing for models on Tera. - [Model Terms](https://docs.tera.gw/legal/model-terms.md): Upstream model licenses and provider terms that apply alongside Tera's Terms of Service. - [Chat completions](https://docs.tera.gw/api-reference/chat-completions.md): OpenAI-compatible chat completions endpoint. Set `stream: true` for Server-Sent Events. Reasoning models return chain-of-thought traces in a separate field — `reasoning` for models using the OpenAI gpt-oss parser (e.g. `openai/gpt-oss-20b`), `reasoning_content` for models using the qwen3 parser (e.g… - [Text completions](https://docs.tera.gw/api-reference/completions.md): OpenAI-compatible legacy completions endpoint. Prefer [chat completions](/api-reference/chat-completions) for new code. - [List models](https://docs.tera.gw/api-reference/models.md): Returns the catalog of models currently available on Tera, including context length, pricing, supported features, and quantization. - [Text to speech](https://docs.tera.gw/api-reference/audio-speech.md): OpenAI-compatible TTS endpoint backed by Kokoro. Returns raw audio bytes. - [Models overview](https://docs.tera.gw/models/overview.md): What's available on Tera today. - [claude-fable-5](https://docs.tera.gw/models/claude-fable-5.md): anthropic/claude-fable-5 on Tera - [claude-opus-4-8](https://docs.tera.gw/models/claude-opus-4-8.md): anthropic/claude-opus-4-8 on Tera - [claude-sonnet-5](https://docs.tera.gw/models/claude-sonnet-5.md): anthropic/claude-sonnet-5 on Tera - [DeepSeek-R1-0528](https://docs.tera.gw/models/deepseek-r1-0528.md): deepseek-ai/DeepSeek-R1-0528 on Tera - [DeepSeek-V3.1](https://docs.tera.gw/models/deepseek-v3-1.md): deepseek-ai/DeepSeek-V3.1 on Tera - [DeepSeek-V3.2](https://docs.tera.gw/models/deepseek-v3-2.md): deepseek-ai/DeepSeek-V3.2 on Tera - [DeepSeek-V4-Flash](https://docs.tera.gw/models/deepseek-v4-flash.md): deepseek-ai/DeepSeek-V4-Flash on Tera - [DeepSeek-V4-Pro](https://docs.tera.gw/models/deepseek-v4-pro.md): deepseek-ai/DeepSeek-V4-Pro on Tera - [gemma-4-26B-A4B-it](https://docs.tera.gw/models/gemma-4-26b-a4b-it.md): google/gemma-4-26B-A4B-it on Tera - [GLM-4.7](https://docs.tera.gw/models/glm-4-7.md): zai-org/GLM-4.7 on Tera - [GLM-5](https://docs.tera.gw/models/glm-5.md): zai-org/GLM-5 on Tera - [GLM-5.1](https://docs.tera.gw/models/glm-5-1.md): zai-org/GLM-5.1 on Tera - [GLM-5.2](https://docs.tera.gw/models/glm-5-2.md): zai-org/GLM-5.2 on Tera - [gpt-5.5](https://docs.tera.gw/models/gpt-5-5.md): openai/gpt-5.5 on Tera - [gpt-5.6-luna](https://docs.tera.gw/models/gpt-5-6-luna.md): openai/gpt-5.6-luna on Tera - [gpt-5.6-sol](https://docs.tera.gw/models/gpt-5-6-sol.md): openai/gpt-5.6-sol on Tera - [gpt-5.6-terra](https://docs.tera.gw/models/gpt-5-6-terra.md): openai/gpt-5.6-terra on Tera - [gpt-oss-120b](https://docs.tera.gw/models/gpt-oss-120b.md): OpenAI's open-weight 120B reasoning model on Tera. OpenAI-compatible. 131k context. US-only, zero retention. - [gpt-oss-20b](https://docs.tera.gw/models/gpt-oss-20b.md): OpenAI's open-weight 20B reasoning model on Tera. OpenAI-compatible. 131k context. US-only, zero retention. - [Kimi-K2.5](https://docs.tera.gw/models/kimi-k2-5.md): moonshotai/Kimi-K2.5 on Tera - [Kimi-K2.6](https://docs.tera.gw/models/kimi-k2-6.md): moonshotai/Kimi-K2.6 on Tera - [kimi-k2-thinking](https://docs.tera.gw/models/kimi-k2-thinking.md): moonshotai/kimi-k2-thinking on Tera - [Kimi-K3](https://docs.tera.gw/models/kimi-k3.md): moonshotai/Kimi-K3 on Tera - [Llama-3.3-70B-Instruct](https://docs.tera.gw/models/llama-3-3-70b-instruct.md): meta-llama/Llama-3.3-70B-Instruct on Tera - [MiniMax-M2](https://docs.tera.gw/models/minimax-m2.md): MiniMaxAI/MiniMax-M2 on Tera - [Qwen3-235B-A22B-Instruct-2507](https://docs.tera.gw/models/qwen3-235b-a22b-instruct-2507.md): Qwen/Qwen3-235B-A22B-Instruct-2507 on Tera - [Qwen3-Coder-480B-A35B-Instruct](https://docs.tera.gw/models/qwen3-coder-480b-a35b-instruct.md): Qwen/Qwen3-Coder-480B-A35B-Instruct on Tera - [Qwen3-Next-80B-A3B-Instruct](https://docs.tera.gw/models/qwen3-next-80b-a3b-instruct.md): Qwen/Qwen3-Next-80B-A3B-Instruct on Tera - [Qwen3-Next-80B-A3B-Thinking](https://docs.tera.gw/models/qwen3-next-80b-a3b-thinking.md): Qwen/Qwen3-Next-80B-A3B-Thinking on Tera ## OpenAPI Specs - [openapi](/openapi.yaml)