Python serverless compute with a Sandbox API for untrusted agent code, scale-to-zero, and per-second CPU/GPU billing.

Daytona is now a closed-source sandbox cloud. GitHub is archived. Starter $20, Plus $200, Scale $500, plus compute after included credits.

GPU rental marketplace with interruptible and on-demand instances. Prices move by host. Confirm the live offer before you rent.

Meta ANN library. A library, not a hosted vector DB or marketing ally.

An efficient local LLM inference library built around the EXL3 quantization format, running large open models on consumer GPUs with tensor and expert parallelism, and powering the TabbyAPI server.

AI-native GPU cloud. No invented seat.

halogen-flash-server is a Strix Halo inference server that runs Qwen3.8-Flash-Next with an OpenAI-compatible API.

NInfer is an Apache-2.0 C++/CUDA engine for selected Qwen checkpoints on a single RTX 5090, with OpenAI- and Anthropic-compatible local APIs.

NVIDIA CUDA MCP server and Nsight Copilot for coding agents. Hosted docs MCP plus an Apache 2.0 self-hosted blueprint.

Inco AI's Apache 2.0 block-diffusion drafter for speculative decoding. Released 2026-08-18 for Qwen3.8-27B and Muse Glimmer.

inclusionAI MIT hybrid-linear MoE: 124B total, 5.1B active.

Ling-3.0-tiny is inclusionAI's MIT hybrid-linear MoE: 7.9B total, 1.3B active, for local agents.

mini-AGI is a byte-level model that trains from scratch on an 8GB GPU and keeps learning from one stream without forgetting what it knew.

YuE2-3B is M·A·P's CC BY-NC music model that plans an editable score, then renders vocals and accompaniment locally.