CNWPlayer/VegaLM1-42M-CodeCompletion
Updated
🌌 It's coming... i think
CNWPlayer
CNWPlayer
AI & ML interests
None yet
Recent Activity
updated a model 13 minutes ago
CNWPlayer/VegaLM1-42M-CodeCompletion published a model about 1 hour ago
CNWPlayer/VegaLM1-42M-CodeCompletionOrganizations
None yet
updated a
model 13 minutes ago
published a
model about 1 hour ago
reacted to dejanseo's post with 🔥 about 2 hours ago
Post
295
Ox Alpha is GLM
https://dejan.ai/blog/ox-alpha/
A parameter-free k-nearest-neighbour classifier over Normalized Compression Distance (Lee et al., MobiSys ’24, Eq. 1, built on Jiang et al.'s gzip-based text classifier). NCD compares two texts by how well they compress together. C(s) is the gzip-compressed length of s. Text sharing an author's patterns compresses better together than text from a different author, so the method needs no model weights and no embeddings.
The reference corpus covers 60 prompts (essays, code, emails, dialogue, poetry) answered by five known models: GPT-5.5, Claude Opus 5, Gemini 3.7 Flash, Gemini 3.1 Pro Preview, and GLM-5.3, for 293 reference texts. ox-alpha answered the first 13 of those prompts, plus one additional novel prompt never given to the reference models beforehand, for 14 queries in total. Each query was classified against the reference corpus independently, with a k-nearest-neighbour vote (k=5):
Model ox-alpha samples matched
GLM-5.3 7 / 14
Claude Opus 5 3 / 14
Gemini 3.7 Flash 2 / 14
GPT-5.5 1 / 14
Gemini 3.1 Pro Preview 1 / 14
GLM-5.3 wins at every k tested: 7/14 at k=3, 7/14 at k=5, 6/14 at k=7, 7/14 at k=9. Claude Opus 5 is the consistent second place.
https://dejan.ai/blog/ox-alpha/
A parameter-free k-nearest-neighbour classifier over Normalized Compression Distance (Lee et al., MobiSys ’24, Eq. 1, built on Jiang et al.'s gzip-based text classifier). NCD compares two texts by how well they compress together. C(s) is the gzip-compressed length of s. Text sharing an author's patterns compresses better together than text from a different author, so the method needs no model weights and no embeddings.
The reference corpus covers 60 prompts (essays, code, emails, dialogue, poetry) answered by five known models: GPT-5.5, Claude Opus 5, Gemini 3.7 Flash, Gemini 3.1 Pro Preview, and GLM-5.3, for 293 reference texts. ox-alpha answered the first 13 of those prompts, plus one additional novel prompt never given to the reference models beforehand, for 14 queries in total. Each query was classified against the reference corpus independently, with a k-nearest-neighbour vote (k=5):
Model ox-alpha samples matched
GLM-5.3 7 / 14
Claude Opus 5 3 / 14
Gemini 3.7 Flash 2 / 14
GPT-5.5 1 / 14
Gemini 3.1 Pro Preview 1 / 14
GLM-5.3 wins at every k tested: 7/14 at k=3, 7/14 at k=5, 6/14 at k=7, 7/14 at k=9. Claude Opus 5 is the consistent second place.
- 1 reply
updated a
model about 14 hours ago
liked a
model 1 day ago
published a
model 1 day ago
reacted to Banaxi-Tech's post with 🔥 1 day ago
Post
3641
We're announcing our BananaMind 2.1 model series!
The models will include:
- BananaMind 2.1 Nano: 10M parameters with 60B tokens.
- BananaMind 2.1 Lite: 25M parameters with 40B tokens.
- BananaMind 2.1 Flash: 50M parameters with 55B tokens.
- BananaMind 2.1 Pro: 135M-145M parameters (still deciding) with 100B tokens.
These model will use a multi tower architecture (like BananaMind/BananaMind-2.1-Unified) with some more architectural changes.
BananaMind 2.1 Pro will probrably use 2 no output towers, instead of one!
We're currently training some experimental models based on this architecture to see its scaling!
Follow us:
BananaMind
@Banaxi-Tech
@vovaRL
@DedeProGames
bananamind-research-community
The models will include:
- BananaMind 2.1 Nano: 10M parameters with 60B tokens.
- BananaMind 2.1 Lite: 25M parameters with 40B tokens.
- BananaMind 2.1 Flash: 50M parameters with 55B tokens.
- BananaMind 2.1 Pro: 135M-145M parameters (still deciding) with 100B tokens.
These model will use a multi tower architecture (like BananaMind/BananaMind-2.1-Unified) with some more architectural changes.
BananaMind 2.1 Pro will probrably use 2 no output towers, instead of one!
We're currently training some experimental models based on this architecture to see its scaling!
Follow us:
@Banaxi-Tech
@vovaRL
@DedeProGames
- 10 replies
replied to Banaxi-Tech's post 1 day ago
eeee! bananas are so yummy and I bet these benchmark numbers will be too
reacted to Enderchef's post with 🔥 2 days ago
Post
4041
Do you support the SLM(1M-150M parameter) community?
If so, join the SLM discord(https://discord.gg/BBYaERvvn), and give these orgs some follows!
BananaMind
SupraLabs
fromziro
AxiomicLabs
If so, join the SLM discord(https://discord.gg/BBYaERvvn), and give these orgs some follows!
- 3 replies
Help training drafters for other models?
#5 opened 3 days ago
by CNWPlayer
Flawless tool calling
🔥 1
2
#6 opened 7 days ago
by CNWPlayer
upvoted a collection 4 days ago
liked a
model 4 days ago
reacted to ProCreations's post with 🔥👍 4 days ago
Post
3366
I have hit 300 followers, and I think this calls for a bit of a giveaway 👀 a unique one, too.
I have had countless AI projects I have wanted to make but have been (brutally) blocked by compute. Now that I finally have just enough compute to sort of get around (i still don't have enough 😭) and for hitting 300 followers (tysm!) I will be funding three of the communities projects via HuggingFace jobs, giving them 150 dollars max worth of compute each. I will personally be picking the winners, I am looking for projects that genuinely hit the compute wall: great ideas, blocked by compute, just like the countless ideas I've had.
To join, head over to https://giveaway.ssh.codes
RULES:
- Final result must be open weight or open source
- Only one submission per person
- Have fun!
I have had countless AI projects I have wanted to make but have been (brutally) blocked by compute. Now that I finally have just enough compute to sort of get around (i still don't have enough 😭) and for hitting 300 followers (tysm!) I will be funding three of the communities projects via HuggingFace jobs, giving them 150 dollars max worth of compute each. I will personally be picking the winners, I am looking for projects that genuinely hit the compute wall: great ideas, blocked by compute, just like the countless ideas I've had.
To join, head over to https://giveaway.ssh.codes
RULES:
- Final result must be open weight or open source
- Only one submission per person
- Have fun!
- 35 replies
replied to ProCreations's post 4 days ago
If it wasn't clearly AI-generated slop that someone's french openclaw agent decided to lay on us then maybe it would have been taken seriously eh?
liked a
model 4 days ago
schizoposting, I'm not the one who shipped anything
replied to ProCreations's post 5 days ago
just report him for being a snob
Ah, monsieur. You've open-sourced your "trainer" for tiny language models, and now you want applause. Let me tell you what you've actually built — and what you're pretending it is.
...
"Claude, roast this person like you're a French critic make no mistakes"