HackerReposA FIELD GUIDE TO OPEN SOURCE
ProSponsor the shelf♧ 0
← Back to top all-time
georgian-io
LLM-Finetuning-Toolkit
Python2023–2024
georgian-io / LLM-Finetuning-Toolkit

LLM-Finetuning-Toolkit

Toolkit for fine-tuning, ablating and unit-testing open-source LLMs.

2Distinct people
4Mentions
0Explicit recommenders
872GitHub stars
REPOSITORY AT A GLANCE

What to know before exploring

What it is

Toolkit for fine-tuning, ablating and unit-testing open-source LLMs.

GitHub

Where it is used

No documented product use has been verified yet.

What people are saying

No recommendation phrase was found in the indexed comments.

No classified warning is attached to this page. Read the discussions before treating silence as approval.

Also in these discussions

No other repository is repeatedly linked in the same discussions.

This week on Hacker News

Nov 9–Nov 16 UTC · Direct repository links and separately labeled related project stories. Indexed through 11/16/2026, 2:19:09 PM UTC.

No direct repository links were found in the indexed HN discussions over the past seven days. Older discussions are available below.

Mentions over time

HN mentions by week4 dated mentions

Observed mentions: 2023-09-04 15:16:42 UTC to 2024-04-07 16:33:19 UTC. Weeks start Monday, UTC.

0122023-09-04 00:00:00 UTC to 2023-09-11 00:00:00 UTC (exclusive): 2 mentions; 2 submissions, 0 commentsSep 42023-09-11 00:00:00 UTC to 2023-09-18 00:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2023-09-18 00:00:00 UTC to 2023-09-25 00:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2023-09-25 00:00:00 UTC to 2023-10-02 00:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2023-10-02 00:00:00 UTC to 2023-10-09 00:00:00 UTC (exclusive): 1 mentions; 1 submissions, 0 comments2023-10-09 00:00:00 UTC to 2023-10-16 00:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2023-10-16 00:00:00 UTC to 2023-10-23 00:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2023-10-23 00:00:00 UTC to 2023-10-30 00:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2023-10-30 00:00:00 UTC to 2023-11-06 00:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2023-11-06 00:00:00 UTC to 2023-11-13 00:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2023-11-13 00:00:00 UTC to 2023-11-20 00:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2023-11-20 00:00:00 UTC to 2023-11-27 00:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2023-11-27 00:00:00 UTC to 2023-12-04 00:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2023-12-04 00:00:00 UTC to 2023-12-11 00:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2023-12-11 00:00:00 UTC to 2023-12-18 00:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2023-12-18 00:00:00 UTC to 2023-12-25 00:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 commentsDec 182023-12-25 00:00:00 UTC to 2024-01-01 00:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2024-01-01 00:00:00 UTC to 2024-01-08 00:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2024-01-08 00:00:00 UTC to 2024-01-15 00:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2024-01-15 00:00:00 UTC to 2024-01-22 00:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2024-01-22 00:00:00 UTC to 2024-01-29 00:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2024-01-29 00:00:00 UTC to 2024-02-05 00:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2024-02-05 00:00:00 UTC to 2024-02-12 00:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2024-02-12 00:00:00 UTC to 2024-02-19 00:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2024-02-19 00:00:00 UTC to 2024-02-26 00:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2024-02-26 00:00:00 UTC to 2024-03-04 00:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2024-03-04 00:00:00 UTC to 2024-03-11 00:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2024-03-11 00:00:00 UTC to 2024-03-18 00:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2024-03-18 00:00:00 UTC to 2024-03-25 00:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2024-03-25 00:00:00 UTC to 2024-04-01 00:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2024-04-01 00:00:00 UTC to 2024-04-08 00:00:00 UTC (exclusive): 1 mentions; 1 submissions, 0 commentsApr 1

Counts come from the repository’s dated source evidence. Index cutoff: 2026-11-02 08:16:59 UTC.

View mention counts and date ranges
UTC ranges include their start and exclude their end. Select a period to filter the source evidence.
PeriodStart (UTC)End (exclusive, UTC)MentionsSubmissionsComments
Sep 42023-09-04 00:00:00 UTC2023-09-11 00:00:00 UTC220
Sep 112023-09-11 00:00:00 UTC2023-09-18 00:00:00 UTC000
Sep 182023-09-18 00:00:00 UTC2023-09-25 00:00:00 UTC000
Sep 252023-09-25 00:00:00 UTC2023-10-02 00:00:00 UTC000
Oct 22023-10-02 00:00:00 UTC2023-10-09 00:00:00 UTC110
Oct 92023-10-09 00:00:00 UTC2023-10-16 00:00:00 UTC000
Oct 162023-10-16 00:00:00 UTC2023-10-23 00:00:00 UTC000
Oct 232023-10-23 00:00:00 UTC2023-10-30 00:00:00 UTC000
Oct 302023-10-30 00:00:00 UTC2023-11-06 00:00:00 UTC000
Nov 62023-11-06 00:00:00 UTC2023-11-13 00:00:00 UTC000
Nov 132023-11-13 00:00:00 UTC2023-11-20 00:00:00 UTC000
Nov 202023-11-20 00:00:00 UTC2023-11-27 00:00:00 UTC000
Nov 272023-11-27 00:00:00 UTC2023-12-04 00:00:00 UTC000
Dec 42023-12-04 00:00:00 UTC2023-12-11 00:00:00 UTC000
Dec 112023-12-11 00:00:00 UTC2023-12-18 00:00:00 UTC000
Dec 182023-12-18 00:00:00 UTC2023-12-25 00:00:00 UTC000
Dec 252023-12-25 00:00:00 UTC2024-01-01 00:00:00 UTC000
Jan 12024-01-01 00:00:00 UTC2024-01-08 00:00:00 UTC000
Jan 82024-01-08 00:00:00 UTC2024-01-15 00:00:00 UTC000
Jan 152024-01-15 00:00:00 UTC2024-01-22 00:00:00 UTC000
Jan 222024-01-22 00:00:00 UTC2024-01-29 00:00:00 UTC000
Jan 292024-01-29 00:00:00 UTC2024-02-05 00:00:00 UTC000
Feb 52024-02-05 00:00:00 UTC2024-02-12 00:00:00 UTC000
Feb 122024-02-12 00:00:00 UTC2024-02-19 00:00:00 UTC000
Feb 192024-02-19 00:00:00 UTC2024-02-26 00:00:00 UTC000
Feb 262024-02-26 00:00:00 UTC2024-03-04 00:00:00 UTC000
Mar 42024-03-04 00:00:00 UTC2024-03-11 00:00:00 UTC000
Mar 112024-03-11 00:00:00 UTC2024-03-18 00:00:00 UTC000
Mar 182024-03-18 00:00:00 UTC2024-03-25 00:00:00 UTC000
Mar 252024-03-25 00:00:00 UTC2024-04-01 00:00:00 UTC000
Apr 12024-04-01 00:00:00 UTC2024-04-08 00:00:00 UTC110

What Hacker News says

1 sources
rohit_saha·Oct 8, 2023Submission

Show HN: Finetuning LLMs: Open-source vs. Close-source

Hello all, I have been working on benchmarking different LLMs -- both open-source and closed-source. Repo: https://github.com/georgian-io/LLM-Finetuning-Hub Precisely, I am comparing their out-of-the-box capabilities (prompting) and their fine-tuned conterparts! So far, the following models have been benchmarked: Open-Source: - Flan-T5-large (780M) - RedPajama (3B & 7B) - Falcon 7B - Llama2 (7B & 13B) - Mistral 7B - Mosaic MPT 7B Close-Source: - AI21 Jurassic-2 (Light, Grande, Jumbo) - Writer Palmyra 30B - GPT 3.5 Turbo 154B The following trends have emerged: - For out-of-the-box zero-shot & few-shot prompting, GPT-3.5-turbo takes the cake! Highly likely that being the biggest model out of all helps with generalizability! - Even among other closed-source LLMs such as Jurassic-2 and Palmyra, GPT-3.5-turbo wins! - Open-source models, howeve
Read the rest
r, do not fare very well for out-of-the-box tasks. Between all open-source models, Mistral-7B achieves the best stats! - When it comes to finetuning, things get very competitive! We notice that much smaller models such as Llama2-7B, Mistral-7B, Falcon-7B are able to compete with the likes of GPT-3.5-turbo 154B and Jurassic-2 models. The last point makes me very hopeful that smaller LLMs when finetuned on narrow use-cases / data can give a tough competition to the much larger models. I am aiming to benchmark more models including Anthropic Claude2, Google PaLMv2, CoHere Command and Inflection Pi. My hunch is these huge closed-source models will generally perform well out-of-the-box compared to (smaller) open-source models. However, finetuning will change the game, where smaller models will compete or even out-compete the larger models! Since there are so many LLMs out there, would love to get some help, in case anyone consider contributing :)
2 points · Open discussion

MAKE YOURSELF AT HOME

A few shortcuts

/
Search the collection
?
Open this guide
Esc
Close dialogs and navigation

Your saved repositories and interests stay in this browser.