Skip to content

Pull requests: openai/parameter-golf

New pull request New
300 Open 112 Closed
300 Open 112 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

Radial bitnet submission
#435 opened Mar 22, 2026 by rthgit Loading…
10L XSA + LeakyReLU² + Partial RoPE (val_bpb=1.1370)
#434 opened Mar 22, 2026 by parinzee Loading…
EBLS Learned Sharing (10min/16MB)
#433 opened Mar 22, 2026 by Robby955 Loading…
Add non-record 1x5090 autoresearch submission with two-campaign analysis
#432 opened Mar 22, 2026 by jadechip Loading…
Add non-record 1xH100 auto precision budget experiment
#431 opened Mar 22, 2026 by spatnala18 Loading…
TrigramHash + XSA + TTT on thwu1 SOTA stack - val_bpb pending H100
#430 opened Mar 22, 2026 by sahiee-dev • Draft
Non-record: 11L EMA + GPTQ-lite + warmdown3500 + QAT@0.15 control (val_bpb=1.1231, 8xH100 verified)
#429 opened Mar 22, 2026 by AbhisekBasu1 Loading…
AnAI: BigramHash(12288) + TrigramHash + int5/int6 QAT
#427 opened Mar 22, 2026 by VeerGosai Loading…
4 tasks
Record: 10L Int5-MLP + Mixed Quant + GradClip + Warmdown3k (mean val_bpb=1.20262)
#426 opened Mar 22, 2026 by aniketio-ctrl Loading…
5 tasks done
Codex/runpod 2026 03 20 checkpoint
#425 opened Mar 22, 2026 by DeividasMat Loading…
Add non-record EMA and adaptive export exploration
#424 opened Mar 22, 2026 by someone114514 Loading…
efficient recurrent gpt
#423 opened Mar 22, 2026 by GITHUB-SUBASHk Loading…
Record: 11L GradQuant + EMA + Sliding Eval (val_bpb=1.1416)
#422 opened Mar 22, 2026 by albertorkive Loading…
Non-record: 11L mixed int5/int6 + working QAT + TTT (val_bpb=1.1466)
#421 opened Mar 22, 2026 by vytautas-bunevicius Loading…
2 of 4 tasks
WIP: Shared-transformer + warmdown-aligned training (not final submis…
#420 opened Mar 22, 2026 by leofeasby Loading…
feat: Add Andreas' first test submission
#419 opened Mar 22, 2026 by AndreasBis Loading…
6
Non-record: PrismLM v3 — DiffTransformer V2 + NorMuon + TrigramHash (val_bpb=1.1715)
#418 opened Mar 22, 2026 by yashverms Loading…
4 of 6 tasks
Record: 11L XSA4 + Tight SWA + FA3 + Two-Phase TTT (3-seed mean val_bpb=1.1227)
#417 opened Mar 22, 2026 by EthanYangTW Loading…
9
Add non-record 11L XSA4 EMA run (val_bpb 1.12296, over 16MB)
#416 opened Mar 22, 2026 by kshitizz36 Loading…
Record: 11L EMA + GPTQ-lite + warmdown3500 + QAT@0.15 (val_bpb=1.1233)
#414 opened Mar 22, 2026 by signalrush Loading…
6 tasks done
Non-record: Value Residual (-0.015 BPB) + Gated Attention (-0.003 BPB) with ablations
#413 opened Mar 22, 2026 by anantdgoel Loading…
WIP: Shared depth training
#412 opened Mar 22, 2026 by tranphat180603 Loading…
First submission
#408 opened Mar 22, 2026 by markste-in Loading…
Add non-record 1xH200 fp16-embed baseline sweep submission
#407 opened Mar 22, 2026 by itu-itis24-buyukhelvacigilm24 Loading…
Non-record: 11L XSA4 + EMA + SDTTT (3-seed mean val_bpb=1.1287)
#406 opened Mar 22, 2026 by dentity007 Loading…
5 tasks done
ProTip! Mix and match filters to narrow down what you’re looking for.