Progressive UI from LLM

O(n) streaming JSON parser for LLM tool calls. Agents act sooner, abort bad outputs early. WASM SIMD, up to 2000× faster than stock AI SDK parsers.

Parse partial JSON as it streams in. Type-safe, tiny, built for LLM output.

A utility for efficient, schema-validated parsing of JSON streams in real-time.

Streaming JSON parser for LLM outputs. Zod-typed partials, zero runtime deps. Native adapters for OpenAI, Anthropic, Ollama, and raw fetch/SSE.

Streaming JSON for Web Streams: NDJSON, incremental array streaming, and partial JSON for LLM output. Zero dependencies, runs on every runtime.

Best-effort parser for incomplete JSON. Render streaming LLM output before it finishes arriving. Zero dependencies, never throws.

A React library for building generative UIs from streaming LLM / backend responses.

Best-effort parsing of partial/streaming JSON from LLM tool-calls — zero dependencies

Stream typed, validated partial objects out of an LLM token stream — render structured output as it arrives, with per-field completion status.

Extract complete JSON objects from partial, streaming or malformed LLM output — string-literal-aware, LaTeX-safe, zero-dependency TypeScript.

Turn a language model's token stream into a stream of the structured object as it fills in.

Incrementally parse partial / streaming JSON (LLM token streams, SSE) — best-effort object from incomplete JSON, and emit array items / OpenAI deltas as they complete. Stdlib only.