atzamis

This makes virtualbox works in a portable mode:

if you have any installed virtualbox then uninstall it first
download https://download.virtualbox.org/virtualbox/7.0.8/VirtualBox-7.0.8-156879-Win.exe or any newer or older version
click on Download ZIP above to download the scripts needed
create a folder inside your USB/external drive or whatever, name it myVMBOX (ex. F:\myVMBOX if your USB is in F:\).
inside F:\myVMBOX put the virtualbox exe we just downloaded (VirtualBox-7.0.8-156879-Win.exe) and rename it to VirtualBox.exe.
inside F:\myVMBOX put the following files start_virtualbox.bat and uninstall_virtualbox.bat, you will find them inside the zip you just downloaded.
now everytime you want to start virtualbox use start_virtualbox.bat.

Which GGUF is right for me? (Opinionated)

Good question! I am collecting human data on how quantization affects outputs. See here for more information: ggml-org/llama.cpp#5962

In the meantime, use the largest that fully fits in your GPU. If you can comfortably fit Q4_K_S, try using a model with more parameters.

	# train_grpo.py
	#
	# See https://github.com/willccbb/verifiers for ongoing developments
	#
	import re
	import torch
	from datasets import load_dataset, Dataset
	from transformers import AutoTokenizer, AutoModelForCausalLM
	from peft import LoraConfig
	from trl import GRPOConfig, GRPOTrainer

	<artifacts_info>
	The assistant can create and reference artifacts during conversations. Artifacts are for substantial, self-contained content that users might modify or reuse, displayed in a separate UI window for clarity.

	# Good artifacts are...
	- Substantial content (>15 lines)
	- Content that the user is likely to modify, iterate on, or take ownership of
	- Self-contained, complex content that can be understood on its own, without context from the conversation
	- Content intended for eventual use outside the conversation (e.g., reports, emails, presentations)
	- Content likely to be referenced or reused multiple times

	import logging
	import requests
	import time
	from typing import Union, Dict
	from config.settings import get_settings


	class DocumentIntelligenceService:
	"""
	A service class for interacting with Azure Document Intelligence API.

	from typing import Any, Dict, List, Type

	import instructor
	from anthropic import Anthropic
	from config.settings import get_settings
	from openai import OpenAI
	from pydantic import BaseModel, Field


	class LLMFactory: