# Cogito

> Open-source LLM inference, frontier speed. Llama, Qwen, DeepSeek, Mistral on AWS Trainium and NVIDIA GPUs.

## Docs

- [Create chat completion](https://docs.cogito.decart.ai/api-reference/chat-completions.md): POST /v1/chat/completions
- [Errors](https://docs.cogito.decart.ai/api-reference/errors.md): Deterministic error schema. Always includes a request_id.
- [List models](https://docs.cogito.decart.ai/api-reference/list-models.md): GET /v1/models
- [Authentication](https://docs.cogito.decart.ai/getting-started/authentication.md): API keys, header format, and best practices.
- [Model catalog](https://docs.cogito.decart.ai/getting-started/models.md): Available open-source models, with context windows, throughput, and pricing.
- [Overview](https://docs.cogito.decart.ai/getting-started/overview.md): Cogito is open-source LLM inference, served on AWS Trainium and NVIDIA GPUs. OpenAI-compatible from line one.
- [Pricing](https://docs.cogito.decart.ai/getting-started/pricing.md): How tokens are counted, billed, and discounted.
- [Quickstart](https://docs.cogito.decart.ai/getting-started/quickstart.md): Send your first Cogito request in under five minutes.
- [Function calling](https://docs.cogito.decart.ai/guides/function-calling.md): Wire LLMs into your tools with reliable, parallel function calls.
- [Streaming](https://docs.cogito.decart.ai/guides/streaming.md): SSE chunks, cancellation, error recovery.
- [Structured outputs](https://docs.cogito.decart.ai/guides/structured-outputs.md): Guaranteed JSON schema adherence on every catalog model.
- [DeepSeek V3.2](https://docs.cogito.decart.ai/models/deepseek-v3-2.md): The proven workhorse. Strong reasoning at one of the lowest prices in the catalog.
- [DeepSeek V4 Flash](https://docs.cogito.decart.ai/models/deepseek-v4-flash.md): Cheap, fast, 1M context. The high-volume workhorse.
- [DeepSeek V4 Pro](https://docs.cogito.decart.ai/models/deepseek-v4-pro.md): Frontier-class reasoning with a 1M-token context window.
- [GPT-OSS 120B](https://docs.cogito.decart.ai/models/gpt-oss-120b.md): OpenAI's open-weight 120B MoE. Strong general reasoning at near-commodity pricing.
- [Kimi K2.6](https://docs.cogito.decart.ai/models/kimi-k2-6.md): Moonshot's flagship MoE for coding agents and long-horizon reasoning.

## Optional

- [Platform](https://cogito.decart.ai)
- [GitHub](https://github.com/DecartAI)