Code

Sn1 437 implement gemma 3 27 b it#644

Code

Merged

bkb2135 merged 24 commits intostagingmacrocosm-os/prompting:stagingfrom SN1-437-implement-gemma-3-27-b-itmacrocosm-os/prompting:SN1-437-implement-gemma-3-27-b-itCopy head branch name to clipboardMar 18, 2025

Commits

Commits on Mar 12, 2025

• Merge pull request #621 from macrocosm-os/staging

  Show description for 9f90feb
  bkb2135

  authored
  9f90feb

  Copy full SHA for 9f90feb

• Merge pull request #642 from macrocosm-os/staging

  Show description for 9a53570
  bkb2135

  authored
  9a53570

  Copy full SHA for 9a53570

Commits on Mar 14, 2025

• Add gemma model and change llm_model to list

  bkb2135

  committed
  7aa9052

  Copy full SHA for 7aa9052

• Linting

  bkb2135

  committed
  b992a27

  Copy full SHA for b992a27

• Remove class method

  bkb2135

  committed
  44001c2

  Copy full SHA for 44001c2

• Adjust for LLM_MODEL being a list

  bkb2135

  committed
  071e725

  Copy full SHA for 071e725

• Add logic for deleting hf models

  bkb2135

  committed
  4fa07ce

  Copy full SHA for 4fa07ce

• Move to cpu then delete

  bkb2135

  committed
  1904ff7

  Copy full SHA for 1904ff7

Commits on Mar 17, 2025

• Fix multi-choice

  dbobrenko

  committed
  10f5897

  Copy full SHA for 10f5897

• Update version

  dbobrenko

  committed
  a164021

  Copy full SHA for a164021

• v2.17.6

  Show description for 6f9dbb7
  dbobrenko

  authored
  6f9dbb7

  Copy full SHA for 6f9dbb7

• Merge branch 'main' of https://github.com/macrocosm-os/prompting into SN1-437-implement-gemma-3-27-b-it

  bkb2135

  committed
  f3ea57e

  Copy full SHA for f3ea57e

• Remove unused pipeline

  bkb2135

  committed
  6142366

  Copy full SHA for 6142366

• Load model based on settings config

  bkb2135

  committed
  2280b6f

  Copy full SHA for 2280b6f

• Update pyproject.toml

  bkb2135

  committed
  7eacb91

  Copy full SHA for 7eacb91

• Rename llama key

  bkb2135

  committed
  e8b18aa

  Copy full SHA for e8b18aa

• Update Lock File

  bkb2135

  committed
  4be892e

  Copy full SHA for 4be892e

• Format gemma messages before inferencing

  bkb2135

  committed
  b3151bd

  Copy full SHA for b3151bd

• Create 3 classes

  bkb2135

  committed
  4610cd4

  Copy full SHA for 4610cd4

• Alternate Assistant and User

  bkb2135

  committed
  4ca49f7

  Copy full SHA for 4ca49f7

Commits on Mar 18, 2025

• Precommit Fix

  richwardle

  committed
  03d1633

  Copy full SHA for 03d1633

• Update Gemma Syntax

  bkb2135

  committed
  f7b8b45

  Copy full SHA for f7b8b45

• Replace Gemma with Mistral Model

  bkb2135

  committed
  fbf3611

  Copy full SHA for fbf3611

• Simplify Unload Model

  richwardle

  committed
  25b9586

  Copy full SHA for 25b9586