Slotstream runs AI models too large for your Mac's memory by keeping shared weights resident and streaming mixture-of-experts layers from SSD on demand. A single native Swift binary built on MLX and Metal serves Ollama, OpenAI and Anthropic compatible APIs, with one command to launch Claude Code against the local model. It targets Apple Silicon Macs with 16 to 64 GB of RAM and about 110 GB of free disk; a 48 GB machine reports around 21 tokens per second on a 105 GB model.
Be the first to rate Slotstream. Your feedback helps other users decide.
Run a 105 GB model on 16 to 64 GB of RAM through expert streaming
Launch Claude Code, Codex or opencode against the model with one command
Persist long conversations and shared system prompts across restarts
Similar tools in the same category — compare side-by-side
OpenAPPA is a deterministic guardrail engine for AI agents that tracks data flow instead of matching...
ai-rete-RAG pairs a Rete rule engine with retrieval-augmented generation so a decision stays determi...
Scry is a programmable search MCP server that lets an agent run one bounded query across 108 billion...
Guides and comparisons about code tools
A roundup of 18 AI tools launched between August and October 2026 that reached Hacker News front page: coding agents, guardrails, on-device models, dictation.
Nine Show HN launches beneath the model: filesystems, snapshots, gateways and deploy tools that decide what an agent sees and what survives.
Nine launches that check agent work: per-call access policy, MCP server vetting, runtime verification, readable diffs and a hard stop at 95%.
Nine Show HN launches sitting under the agent: a model gateway, vectorless retrieval, on-device inference, an MCP browser layer and connectors for real systems.
Eight Show HN projects for trusting long agent runs: work context capture, distilled memory, folder fences, dollar budgets, plan review and tool governance.
