Models

Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution

StepFun's next-generation flagship base model for coding and professional knowledge work, with native text, image, and video input and a 1M-token context window

Multimodal Kimi model with 1M context and toggleable max-effort thinking for long-horizon agent work

Efficient model for low-latency assistance, extraction, and routine automation

October 17 update of DeepSeek V4.1 Flash with improved reasoning, coding, agentic tasks, and visual understanding

MiMo Flash model for multimodal coding agents and long-context automation

DeepSeek V4 Pro snapshot with million-token context and support for thinking and non-thinking modes

Experimental multimodal DeepSeek V4 Flash model for image understanding, coding, and agentic work

Official DeepSeek V4 Flash release with enhanced agentic capabilities and integrated DSpark speculative decoding

Open MoE flagship with million-token context for coding and long agent runs

Multimodal Kimi workhorse for agent loops, coding tasks, and visual context

Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution

Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking

Multimodal MoE reasoning model (276B total, 12B active) for text, image, and audio

Tencent Hy reasoning model for coding, instruction following, and agent tasks

Multimodal MoE reasoning model (975B total, 41B active) for text, image, and audio

DeepSeek V4.1 Flash model for reasoning and agentic coding

Native multimodal hybrid MoE model for image and video understanding, reasoning, and visual agents

Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work

Earlier Kimi frontier model for long-context agents, coding, and multimodal work

Largest Nemotron 3 model for maximum open-weight reasoning and agent accuracy

Thinking Kimi model for slower research passes, planning, and hard technical questions

Open MiMo model for multimodal coding agents and long-context automation

Hybrid-reasoning DeepSeek model with thinking and non-thinking modes, sparse attention, and tool-use

Hybrid-linear MoE reasoning model with 124B total and 5.1B active parameters for efficient agents

Muse Glimmer is a 30-billion-parameter open-weight multimodal model from Meta Superintelligence Labs, distilled from Muse Spark for always-on local agents, tool use, coding, and image understanding.

StepFun flash lane for quick multimodal reasoning and coding assistance

Open Gemma instruction model for efficient chat and self-hosted deployments

Largest Gemma 4 instruction model for open, self-hosted chat and reasoning

DeepSeek chat model for instruction following, coding, and analysis

May 2025 DeepSeek R1 update with improved mathematics, coding, and reasoning

Fast NVIDIA Nemotron MoE for reliable agentic tasks across enterprise workloads

Nemotron middle tier for collaborative agents and high-volume reasoning workloads

Kimi model for long-context chat, coding, and agentic reasoning

Open GPT reasoning model for self-hosted agents and controllable deployments

Reasoning-optimized 398B MoE agent model with extended thinking for long-horizon and multi-turn tool use

Open GPT reasoning model for self-hosted agents and controllable deployments

Small Nemotron 3 MoE for efficient coding, math, and long-context agents

Efficient model for low-latency assistance, extraction, and routine automation

Largest open Gemma 3 instruction model for multilingual text generation and visual understanding

Open multimodal Gemma instruction model for efficient text generation and image understanding

Open multimodal Gemma instruction model for multilingual text generation and image understanding