👽 Big things for X3
Dan P
Datdanboi25
AI & ML interests
Axiomic Labs Founder
Mechatronics engineer, LLMs, Vision Models, Embedding Models
Recent Activity
repliedto Banaxi-Tech's post 33 minutes ago
hi everyone
we have released nacr
its not just any model, its nacr
we have 6 more features and this model only uses 20% of its total capacity!
check it out at https://huggingface.co/saicr/nacr
we're currently working on expanding access as we do more research but right now you have to use our gated access form
follow https://huggingface.co/saicr if you're interested
if you want to join saicr, first read the entire nacr readme, then press the join button.
liked a dataset about 1 hour ago
fromziro/Rephrased2JSON-Preview repliedto their post about 3 hours ago
ForgePlex-M1-6M first model trained on AxiomicLabs TrainWork
https://huggingface.co/ForgeWorks/ForgePlex-M1-6M just dropped from https://huggingface.co/ForgeWorks, and is the first model to ever be trained on our TrainWork training framework.
Achieving an Intelligence Index of 6.87 and taking #22 in the <10m category on the https://huggingface.co/spaces/AxiomicLabs/Open_SLM_Leaderboard, very impressive work for a first model.
Give it some love!Organizations
i got it
9
#1 opened about 9 hours ago
by 
Banaxi-Tech
Add NACR 1.0 and add SAICR organization
1
#166 opened about 9 hours ago
by Banaxi-Tech
Add Hummingbird-V2
😎🚀 1
4
#164 opened 1 day ago
by 
juinron
Add tinctura-v1 (96.2M, from scratch, Index 20.77)
#165 opened about 21 hours ago
by 
TobiasLogic
add ForgePlex-M1-6M
#163 opened 1 day ago
by 
Jsandero
add sol milkshake
3
#160 opened 4 days ago
by 
j0no12
Only PR's will be added for the next little while
👍 3
#162 opened 1 day ago
by 
Datdanboi25
Encouraging entry into CodeSoft's SLM Arena
😔 1
2
#161 opened 4 days ago
by 
wayneworkman2012
[BUG] Kiyo is broken
2
#17 opened 3 days ago
by 
ereniko
GPT-X3
13
#2 opened 3 days ago
by Datdanboi25
Add <1M filter
1
#152 opened 9 days ago
by Banaxi-Tech
What was the peak learning rate used?
#8 opened 4 days ago
by Datdanboi25
Add Pollock 1.5 (127.57M, from scratch, Index 12.59)
2
#159 opened 4 days ago
by 
dawidmajewski
Update cagliostro-v3 to final weights (75B tokens)
2
#158 opened 5 days ago
by TobiasLogic
Add cagliostro-v3 (146M, from scratch, Index 26.27)
#157 opened 5 days ago
by TobiasLogic
Submission: cagliostro-v3 (146M, from scratch, Index 26.27)
2
#156 opened 5 days ago
by TobiasLogic
Update README.md
#2 opened 5 days ago
by Datdanboi25
Add peacebell-v1-148M (148.55M, WWII-only, from scratch)
#155 opened 5 days ago
by wayneworkman2012
invalid discord invite
3
#5 opened 5 days ago
by j0no12
Add JugnuLM-110M-R2+ (altslate) Sub-150M, pretrained from scratch (Qwen3-arch + value residuals + Muon, WSD schedule, ~25B tokens on FineWeb-Edu). Open training checkpoint (not merged): https://huggingface.co/altslate/JugnuLM-110M-R2plus Zero-shot, LM-eval harness (acc_norm), + AxiomicLabs ArithMark-3 script: arc (ARC-Easy) 47.64 | hellaswag 36.74 | piqa 65.29 | arcChall (ARC-Challenge) 26.96 | arithmark-3 38.60 109,737,302 params. (arithmark-2 omitted — dataset not publicly available.) NOTE for verification: this is a value-residual model, so it must be loaded with trust_remote_code=True (the repo ships modeling_jugnu_vr.py + auto_map). Loading it as a stock Qwen3 silently drops the value-residual pathway and understates the scores (~6 pts ARC-Easy, etc.). The ArithMark-3 script loads it correctly by default.
1
#154 opened 6 days ago
by 
rajpdus