# Boundless

> Boundless is the inference partner for AI-native startups, running open-weight models at the price and performance your product deserves.

Boundless (boundless.network) is an AI infrastructure company. Boundless engineers identify the right open-weight model for a customer's workload, tune the serving stack to their traffic, and operate it in production. Customers are AI-native startups whose product depends on inference.

## Pages

- [Home](https://boundless.network/): positioning, approach, and how the engagement works
- [Model library](https://boundless.network/models): every model Boundless serves, open-weight and frontier, with context windows and current pricing
- [Playground](https://inference.boundless.network/playground): try any supported model, one free request
- [News](https://boundless.network/news): announcements and writing from the Boundless team (RSS: https://boundless.network/news/feed.xml)

## Models

- DeepSeek V4 Flash (DeepSeek): best for fast, low-cost general tasks; 1M context
- MiMo-V2.5 (Xiaomi): best for omnimodal understanding across text, image, audio and video; 1M context
- Qwen3.6-35B-A3B (Qwen): best for lightweight workloads at minimal cost; 262K context
- GLM-5.3 Flash (Z.ai): best for low-cost long-context multimodal work at speed; 1M context
- Qwen3.8 Flash (Qwen): best for fast, low-cost general tasks; 1M context
- Nemotron 3 Super (NVIDIA): best for long-context reasoning and tool use; 262K context
- DeepSeek V4.1 Flash (DeepSeek): best for fast, low-cost agentic coding and visual analysis; 1M context
- MiniMax M3 (MiniMax): best for long-horizon agentic work over text, image and video; 1M context
- GPT-5.6 Luna (OpenAI): best for high-volume GPT-5.6 chat, classification, and light agents; 262K context
- DeepSeek V4 Pro (DeepSeek): best for frontier reasoning and agentic coding; 1M context
- GLM-5.2 (Z.ai): best for flagship general-purpose chat and reasoning; 262K context
- GPT-5.6 Sol (OpenAI): best for GPT-5.6 flagship for CLI-heavy coding and long-horizon problem solving; 262K context
- Kimi K3 (Moonshot AI): best for agentic coding, tool use, and long-horizon reasoning; 1M context
- Claude Opus 5 (Anthropic): best for anthropic Opus 5 for demanding coding, review, and long-horizon agents; 262K context
- Claude Fable 5.1 (Anthropic): best for claude Fable 5.1 for agentic coding, front-end generation, and knowledge work; 262K context

## Contact

- Talk to an engineer: https://boundless.network/#get-access