Skip to content

Pull requests: google-deepmind/gemma

New pull request New
101 Open 267 Closed
101 Open 267 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

Fix: quantizations issue for multimodel
#519 opened Jan 29, 2026 by pragnyanramtha Loading…
1
Add sandbox restrictions to FileExplorer to prevent arbitrary file access
#518 opened Jan 25, 2026 by markknoffler Loading…
1
Upgrade GitHub Actions for Node 24 compatibility
#515 opened Jan 22, 2026 by salmanmkc Loading…
Fix tool-call parsing to ignore non-tool JSON (fixes #511)
#512 opened Jan 20, 2026 by markknoffler Loading…
1
DOC: Add more specific GPU examples to System Requirements in README.md #395
#508 opened Jan 15, 2026 by abhishekkamble12 Loading…
1
[Combined PR] Security Hardening, Standardization, and Robustness Improvements
#507 opened Jan 13, 2026 by Ashutosh0x Loading…
Fix calculator tool security vulnerability
#506 opened Jan 12, 2026 by markknoffler Loading…
Fix _decode_bytes to handle invalid UTF-8 gracefully
#505 opened Jan 12, 2026 by markknoffler Loading…
tests: relax RMSNorm/RoPE tolerances
#503 opened Jan 12, 2026 by staticpayload Loading…
2
1
docs(ckpts): Clarify 'quantize' behavior in load_params
#502 opened Jan 10, 2026 by Aman071106 Loading…
feat(data): Add 'axis' support to gm.data.pad
#501 opened Jan 10, 2026 by Aman071106 Loading…
Refactor DPO example to parametrize dataset configuration
#499 opened Jan 9, 2026 by Ashish-Patnaik Loading…
1
Refactor classification example to avoid hardcoded label token IDs
#497 opened Jan 9, 2026 by Ashish-Patnaik Loading…
1
Fix SimulateQuantizedEinsum to use einsum_str instead of module name …
#495 opened Jan 8, 2026 by sriramsowmithri9807 Loading…
docs: convert raw sharding.py URL to markdown link
#494 opened Jan 8, 2026 by shivachethanreddy Loading…
1
fix(data): add input validation for empty prompt/response in make_seq2seq_fields
#493 opened Jan 7, 2026 by Ashutosh0x Loading…
1
Add prerequisites section to README
#492 opened Jan 7, 2026 by varma1221 Loading…
2
Fix: Correct typo in _transformer.py
#490 opened Jan 6, 2026 by Ashutosh0x Loading…
1
Fix: Handle empty prompt arrays in make_seq2seq_fields (#483)
#488 opened Jan 4, 2026 by Ashutosh0x Loading…
1
4
Feat/gemma3 model intelligence
#487 opened Jan 1, 2026 by junaidaslam2006 Loading…
docs: fix subject-verb agreement in README
#486 opened Jan 1, 2026 by gabrielfranca95 Loading…
1
Fix: Handle empty prompt gracefully in make_seq2seq_fields with warning
#484 opened Dec 31, 2025 by markknoffler Loading…
1
feat(examples): add native MCP gateway for Gemma 3 function calling
#482 opened Dec 29, 2025 by junaidaslam2006 Loading…
Fix: Pass einsum_str instead of module name to get_axis_to_reduce_from_einsum_str
#480 opened Dec 28, 2025 by markknoffler Loading…
1
1
Initialize devcontainer configuration file
#478 opened Dec 23, 2025 by raedalmalki3 Loading…
1
ProTip! Adding no:label will show everything without a label.