Instructions to use masahiroid/pyrite-stream with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use masahiroid/pyrite-stream with MLX:
# Make sure mlx-lm is installed # pip install --upgrade mlx-lm # if on a CUDA device, also pip install mlx[cuda] # Generate text with mlx-lm from mlx_lm import load, generate model, tokenizer = load("masahiroid/pyrite-stream") prompt = "Once upon a time in" text = generate(model, tokenizer, prompt=prompt, verbose=True) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- MLX LM
How to use masahiroid/pyrite-stream with MLX LM:
Generate or start a chat session
# Install MLX LM uv tool install mlx-lm # Generate some text mlx_lm.generate --model "masahiroid/pyrite-stream" --prompt "Once upon a time"
- Atomic Chat
File size: 536 Bytes
cab2ef1 21f1f04 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 | """Project Pyrite Phase 2: MoE-Stream Engine.
Decoupling Resident Attention and Dynamic NVMe-Streamed Experts
for 100B+ MoEs on Unified Memory Consumer Hardware.
Author: Masahiro NAKATSUGAWA (中津川 昌弘)
Affiliations: n.p.minerva & Co. / AntenneFrance
"""
from .expert_registry import SafetensorsIndex, ModelSafetensorsIndex
from .moe_stream_layer import StreamingMoELayer, ExpertLRUCache
__version__ = "0.2.0"
__all__ = [
"SafetensorsIndex",
"ModelSafetensorsIndex",
"StreamingMoELayer",
"ExpertLRUCache",
]
|