Release v0.3.3 · PrunaAI/pruna

The juiciest bits 🧃

Algorithm & compatibility improvements ⚙️

feat: FA3-FP8-extension by @Marius-Graml in #552
feat: extend moe model check to multimodal ones and add block quantization parameters by @llcnt in #605
feat: moe kernel tuning by @llcnt in #482

Lots of new tools for benchmarking and evaluation 📊

feat(evaluation): Add lm-eval to Pruna Metrics by @sky-2002 in #380
feat(metrics): DINO Score v3 by @davidberenstein1957 in #568
feat: add column_map support to collate functions by @zamal-db in #561
feat: initial implementation for rapidata by @begumcig in #581
feat: add benchmark support to PrunaDataModule and implement PartiPrompts by @davidberenstein1957 in #502

Python 3.13 and pytorch 2.11 support is here 🐍

fix(evaluation): replace bare raises with proper exceptions and add text_generation_quality request by @zamal-db in #560
fix: protect lm-eval import to allow evaluation-agent import without extra by @gsprochette in #586
fix(torchao): update imports of quantizer by @ParagEkbote in #549
fix: wrap callable enum values with enum.member for python 3.13 by @gsprochette in #583
fix: remove pruna-pro hook from pre-commit by @ParagEkbote in #572
fix: cache handling in SmashConfig due to invalid path exception by @ParagEkbote in #598
fix: pre-download sage_attention kernel before applying backend, remove pinned fa3 kernel version by @Marius-Graml in #578
fix: ping peft >= 0.18.0, < 0.19.0 for torchao compatability issues by @davidberenstein1957 in #630

We’ve made a bunch of improvements to make installing, testing, and developing locally faster and more reliable:

ci: fix too many requests http error in the cpu tests by @begumcig in #577
ci: add uv virtual environment cache to setup-uv-project action by @davidberenstein1957 in #559
ci: separate extra installs by @begumcig in #622
build: make index pypi default and pythonanywhere explicit and setup python through uv by @gsprochette in #613
ci: restrict build to manual dispatch and version tags by @gsprochette in #633
test: explicit combo names and switch stable fast fixture by @gsprochette in #582

We also updated our PR template and Readme for smoother contributions!:

docs: update readme cta by @sdiazlor in #591
docs: updated PR template by @minettekaum in #576
docs: added comment about vibe coded solutions to pr template by @minettekaum in #606

@zamal-db made their first contribution in #560 (and they did not stop and did another contribution right after with #561)
@sky-2002 made their first contribution in #380 and added the entire lm-eval harness to Pruna!

Full Changelog: v0.3.2...v0.3.3