DwarfStar (DS4)

Needle 2 is an open-source, 45-million-parameter foundation model for tool calling, device use, and structured extraction on small devices. Its weights are compressed into a single 14 MB binary and integrated with its inference engine; the repository reports that a full session uses about 28 MB of RAM and performs inference without network access after setup. The model converts text and tool descriptions into schema-constrained JSON calls using a byte-level grammar, attaches a learned confidence score for threshold-based escalation, and can retrieve the top five tools from a larger catalogue. It uses a 256-token sliding window with tools pinned as key-value sinks to bound memory, and its Simple Attention Network combines a Hadamard MLP, grouped-query attention, engram key-value memory, and multi-lane hyper-connections. The Python package supports inference, LoRA fine-tuning, and checkpoint export; the inference engine is fetched from Hugging Face and cached, while the training stack is an optional installation.

mini-AGI is an experimental continual-learning, byte-level language model that trains from scratch on a CUDA-capable GPU with at least 8 GB of VRAM. It reads 256 byte values directly rather than using a conventional tokenizer, processes data one chunk at a time, and uses the same forward path for training and generation. Its architecture combines dense prelude blocks with a recurrent block applied up to 24 times, adaptive PonderNet-style halting, and per-application top-8 routing through a dynamically growing and pruning expert pool. Expert weights and Adam moments are stored as files on disk; a working set is paged through RAM and VRAM as needed, allowing the pool size to exceed available VRAM. The project includes Python commands for building corpora, reading files, streaming training, serving a local web interface, and inspecting or replicating continual-learning experiments. The repository describes the current model as toy-level rather than frontier-capable, and its weights are generated in the local weights directory rather than distributed with the source.

DeepSeek (深度求索) is an AI research project and model provider that develops and open-sources large foundation models (notably named DeepSeek-V4 and DeepSeek-R1) and provides hosted inference via a chat interface (“DeepSeek AI”) and an API. It publishes open-source model releases and offers a hosted API for text-based prompt inference; video reports state DeepSeek models can be exposed as external backends and routed into the Codex model picker via Codex Router.

LLM Space v4 is a local-first desktop workbench for prototyping and developing AI agents. It lets users version prompts, system messages, tools, and model settings; trace each model call and tool run in an agent loop; replay historical runs to debug failures; and evaluate agent performance across runs. It can generate prompts and tools with AI and turn a thread into a runnable LangGraph agent. Threads, project files, and API keys remain on the user's computer, while the project describes the app as cloud-ready for managed agents. The desktop application uses Electrobun with a React, Tailwind CSS, and shadcn/ui interface, and is built as a Bun monorepo around Pi Agent Core.

NVIDIA DGX Spark is a compact AI workstation from NVIDIA for local development, fine-tuning, and inference of generative-AI models. It is built around the NVIDIA GB10 Grace Blackwell Superchip, with 128 GB of unified memory and up to 1 petaflop of FP4 AI performance, and runs NVIDIA DGX OS.