talkie-lm/talkie

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

1015 stars · 60 forks observed · 2026-09-30

What it is AI-extracted, prompt v1, taxonomy v1, 2026-09-07, confidence 0.9

talkie is a Python library and CLI for running inference with the talkie 13B language model family, including a 13B model trained on pre-1931 English text and an instruction-tuned variant. It provides a simple API for downloading models from HuggingFace and generating text completions, streaming tokens, and multi-turn chat.

Use cases

• run inference with a 13B language model trained on pre-1931 text
• generate vintage-style text completions from a Python API
• chat with an instruction-tuned model tuned on etiquette manuals and encyclopedias
• compare a 1930-era model against a modern FineWeb-trained baseline
• stream tokens from a language model in Python
• download and run LLM checkpoints from HuggingFace via CLI

When to choose

• you want a research or novelty LLM trained on historical pre-1931 text
• you need a simple Python API or CLI to run a 13B model on a single large GPU
• you are studying how temporal corpus coverage affects model behavior with a controlled modern baseline

When to avoid

• you need state-of-the-art general-purpose LLM quality
• you lack a CUDA GPU with at least 28 GB VRAM or the ~26-50 GB disk space per model
• you need multi-GPU, quantized, or CPU-only inference

Facets

library · maturity active

llm-inference machine-learning cli large-language-models artificial-intelligence developer-tools python cli language-model inference huggingface research-model vintage-text pytorch natural-language-processing gpu linux macos

Member repositories

For agents

markdown · JSON · MCP: product_card(name="talkie-lm/talkie")