Explore Text Models

Kimi K2.7 Code is Moonshot AI's coding-focused agentic model built for long-horizon software engineering workflows. It supports native image input, tool calling, and forced thinking mode; instant/non-thinking mode is not supported.

Nvidia Nemotron 3 Ultra 550B running through Chutes TEE with encrypted transport and provider evidence support. Not included in the subscription.

MiMo V2.5 Pro UltraSpeed is Xiaomi's speed-focused 1T-parameter MiMo V2.5 Pro mode, built for near-instant coding assistance, real-time chat, live edits, and low-latency agent loops. Xiaomi reports up to roughly 1,000 tokens per second using its TileRT serving stack, FP4 expert quantization, and DFlash speculative decoding.

DeepSeek V4 Flash is an efficiency-optimized Mixture-of-Experts model from DeepSeek with a 1M-token context window. Running inside a TEE (Trusted Execution Environment), with provider attestation support.

Qwen3.6 35B A3B is an open-weight MoE model from Alibaba's Qwen team with 35B total parameters and 3B active parameters per token. Running inside a TEE (Trusted Execution Environment), with provider attestation support.

Linkup Research with high reasoning depth. Runs an async web research agent and returns a sourced answer for complex research tasks. Responses can take several minutes.

Linkup Research with low reasoning depth. Runs an async web research agent and returns a sourced answer for factual and multi-source questions. Responses can take several minutes.

Linkup Research with medium reasoning depth. Runs an async web research agent and returns a sourced answer for factual and multi-source questions. Responses can take several minutes.

Linkup Research with extra-high reasoning depth. Runs Linkup's most thorough async web research mode and returns a sourced answer. Responses can take several minutes.

Nex AGI's open-source agentic reasoning model, post-trained on Qwen3.5-397B-A17B. It is built for agentic coding, software engineering, deep research, tool use, and long-horizon tasks with a 256K context window.

Nvidia's Nemotron 3 Ultra 550B A55B model from the Nemotron 3 family. It uses a hybrid Mamba-Transformer MoE architecture. Provider-specific context limits vary, with the longest current route supporting up to 1M context.

Nvidia's Nemotron 3 Ultra 550B A55B model from the Nemotron 3 family. It uses a hybrid Mamba-Transformer MoE architecture. Provider-specific context limits vary, with the longest current route supporting up to 1M context. Thinking enabled.

Qwen3.6 27B is a dense language model from Alibaba's Qwen team with text and image input, configurable thinking/reasoning behavior, and a native 262K context window. Running inside a TEE (Trusted Execution Environment), with provider attestation support.

MiMo V2.5 Pro with Xiaomi thinking enabled for coding, long-context reasoning, and agentic orchestration.

MiMo V2.5 with Xiaomi thinking enabled. It supports deep reasoning, tool calling, structured outputs, and web search with up to 1M context.

Mistral Code Agent Latest is Mistral's direct API alias for devstral-2512, an agentic coding model built for autonomous software engineering, tool use, and long-running code tasks.

Mistral Code Latest is Mistral's direct API alias for codestral-2508, a low-latency coding model for code generation, completion, fill-in-the-middle workflows, function calling, and structured output.

MiniMax M3 is the non-thinking route for MiniMax's open-weights frontier model, built for coding, agent workflows, tool use, and multimodal understanding from step zero. It keeps native thinking disabled for faster direct answers. MiniMax reports 59.0% on SWE-Bench Pro and 66.0% on Terminal Bench 2.1, with Sparse Attention designed to scale context to 1M. It starts with a 512K context cap on NanoGPT for now.