Gary Blankenship garyblankenship

Stream Ollama (openai) chat completion API on CLI with HTTPie and jq

Explanation

This command sends a request to the Chat Completion API to generate high-level documentation for the file @src/arch.js. The API is configured to use the llama3-gradient model and to respond in Markdown format.

The messages array contains two elements:

The first element is a system message that provides the prompt for the API.
The second element is a user message that specifies the file for which to generate documentation.

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.

llama.cpp feature matrix

See the wiki upstream: https://github.com/ggerganov/llama.cpp/wiki/Feature-matrix

Music APIs and DBs

A collection of music APIs, databases, and related tools.

Laravel's Hidden Cache::grace() Pattern Solves the Stale-While-Revalidate Problem

How to serve cached content instantly while updating it in the background—the pattern major CDNs use, now in your Laravel app

Cache invalidation is one of the two hard problems in computer science, and Laravel's standard cache methods force you into an uncomfortable choice: serve stale data or make users wait for fresh data. The Cache::grace() pattern eliminates this trade-off by implementing the stale-while-revalidate strategy that powers major CDNs and high-performance web applications.

This isn't just another Laravel macro—it's a fundamental shift in how you think about caching. Instead of choosing between speed and freshness, you get both: instant responses for users and background updates to keep data current. The pattern is so effective that it's built into HTTP standards (RFC 5861) and used by services like Cloudflare, but most Laravel developers have never heard of it.

Here's why this single caching pattern can

	# Machine Intelligence Made to Impersonate Characteristics: MIMIC

	# NOTE run this $ conda install -c conda-forge mpi4py mpich to get mpi working
	# accelerate launch --use_deepspeed -m axolotl.cli.train ./config_name_here
	base_model: alpindale/Mistral-7B-v0.2-hf
	base_model_config: alpindale/Mistral-7B-v0.2-hf
	model_type: MistralForCausalLM
	tokenizer_type: LlamaTokenizer
	is_mistral_derived_model: true

	name: Deploy to DigitalOcean

	on:
	push:
	branches:
	- main

	jobs:
	deploy:
	runs-on: ubuntu-latest

	<?php

	$response = Http::post(
	"https://generativelanguage.googleapis.com/v1beta/models/gemini-pro:generateContent?key=" .
	config("services.googleai.key"),
	[
	"contents" => [
	[
	"parts" => [
	[

	#!/bin/sh
	# rename-pictures.sh
	# Author: Justine Tunney <[email protected]>
	# License: Apache 2.0
	#
	# This shell script can be used to ensure all the images in a folder
	# have good descriptive filenames that are written in English. It's
	# based on the Mistral 7b and LLaVA v1.5 models.
	#
	# For example, the following command:

	<?php

	namespace App\Http\Controllers;

	use Illuminate\Http\Request;
	use OpenAI\Laravel\Facades\OpenAI;

	class AssistantController extends Controller
	{
	public function generateAssistantsResponse(Request $request)