Skip to content

Pull requests: openai/parameter-golf

New pull request New
660 Open 316 Closed
660 Open 316 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

Phase 1: Three Optimization Variants
#1023 opened Mar 28, 2026 by DanielSooman Loading…
Non-Record: McGilchrist Register Token — causal cumulative mean + FiLM global context pathway
#1022 opened Mar 28, 2026 by aramdov Loading…
Non-record: MC Dropout ensembling is negative for small LMs
#1021 opened Mar 28, 2026 by abaybektursun Loading…
4 tasks done
Add tiny transformer baseline model
#1020 opened Mar 28, 2026 by VettyCordwainer Loading…
Record: AR Self-Gen GPTQ + XSA-all + BigramHash 3072×112 — val_bpb 1.11473 (3-seed mean)
#1019 opened Mar 28, 2026 by abaybektursun Loading…
agent lab/mar28
#1018 opened Mar 28, 2026 by BenedictChannn Loading…
11L VRL + Parallel Muon + Legal TTT v2 (val_bpb=1.1269, non-record)
#1016 opened Mar 28, 2026 by ADIITJ Loading…
3 tasks
8
Add Parameter Golf submission: Vocab768_LinearPhaseInit_GatedXSA_EMA_…
#1015 opened Mar 28, 2026 by shram86 Loading…
N-gram logit boost + HedgeMixer + score-first TTT
#1014 opened Mar 28, 2026 by haimianbaobao007 Loading…
1
Non-record: S4D-Lin SSM Hybrid — Fixing Why Mamba Failed in Parameter…
#1013 opened Mar 28, 2026 by himanshudongre Loading…
Non-record: JEPA-LM — When Synthetic Success Doesn't Transfer to Real…
#1012 opened Mar 28, 2026 by himanshudongre Loading…
Anubhav ctw submission
#1011 opened Mar 28, 2026 by AnubhavBharadwaaj Loading…
Non-record: VRL + LeakyReLU² + Full SOTA Stack (iteration build)
#1010 opened Mar 28, 2026 by AnubhavBharadwaaj Loading…
12L INT4 bQAT + Value Embeddings — val_bpb 1.1588
#1009 opened Mar 28, 2026 by SoHarshh Loading…
Add non-record unlimited-compute 11L LeakyTTT 16h local RTX 4060 Ti run
#1008 opened Mar 28, 2026 by monkeyKingProgrammer Loading…
Submission/hybrid rwkv token shift
#1007 opened Mar 28, 2026 by dillon-blake Loading…
1.1085 BPB: JEPA + AdamW TTT + Full GPTQ + FA3 + LZMA
#1006 opened Mar 28, 2026 by NewyorkDev Loading…
1
[Non-Record] Extended Compute Scaling Analysis: 1.0853 BPB at 50K steps (11.5 hours) on 4×A100MIG
#1005 opened Mar 28, 2026 by OnlyJundong Loading…
Non-record: 33.6M Int5 GPTQ + Legal s_0-only TTT (val_bpb=1.1182)
#1004 opened Mar 28, 2026 by ibarrajo Loading…
7 tasks done
[WIP] Non-record: Probatum — QAT + LeakyReLU² + EMA baseline
#1003 opened Mar 28, 2026 by techleadershipofanurag Loading…
12L INT4 bQAT + EMA Fix + Deterministic QAT — val_bpb ~1.165
#1002 opened Mar 28, 2026 by SoHarshh Loading…
Non-record: Three Approaches + Lessons Learned (best: 1.1188 BPB)
#1001 opened Mar 28, 2026 by ibarrajo Loading…
5 tasks done
[WIP] Latent LM with TurboQuant 3-bit Quantization
#1000 opened Mar 28, 2026 by Elarwei001 Loading…
2 of 6 tasks
Record: 11L Muon TTT + Entropy-Adaptive Epochs (8×H100) — val_bpb 1.1179 (3-seed mean)
#999 opened Mar 28, 2026 by aamodbhatt Loading…
8 tasks done
Add Conker-5 tandem residual exact experts non-record submission
#998 opened Mar 28, 2026 by asuramaya Loading…
1
ProTip! Updated in the last three days: updated:>2026-03-25.