Technology

1919 readers

548 users here now

Which posts fit here?

Anything that is at least tangentially connected to the technology, social media platforms, informational technologies and tech policy.

Rules

1. English only

Title and associated content has to be in English.

2. Use original link

Post URL should be the original link to the article (even if paywalled) and archived copies left in the body. It allows avoiding duplicate posts when cross-posting.

3. Respectful communication

All communication has to be respectful of differing opinions, viewpoints, and experiences.

4. Inclusivity

Everyone is welcome here regardless of age, body size, visible or invisible disability, ethnicity, sex characteristics, gender identity and expression, education, socio-economic status, nationality, personal appearance, race, caste, color, religion, or sexual identity and orientation.

5. Ad hominem attacks

Any kind of personal attacks are expressly forbidden. If you can't argue your position without attacking a person's character, you already lost the argument.

6. Off-topic tangents

Stay on topic. Keep it relevant.

7. Instance rules may apply

If something is not covered by community rules, but are against lemmy.zip instance rules, they will be enforced.

Companion communities

!globalnews@lemmy.zip
!interestingshare@lemmy.zip

Icon attribution | Banner attribution

If someone is interested in moderating this community, message @brikox@lemmy.zip.

founded 1 year ago

MODERATORS

BrikoX@lemmy.zip

New Open Source AI model beats DeepSeek's performance using just 14% of the data its Chinese competitor needed (decrypt.co)

submitted 4 days ago by Hotznplotzn@lemmy.sdf.org to c/technology@lemmy.zip

8 comments fedilink hide all child comments

cross-posted from: https://lemmy.sdf.org/post/29607342

Archived

Here is the data at Hugging Face.

A team of international researchers from leading academic institutions and tech companies upended the AI reasoning landscape on Wednesday with a new model that matched—and occasionally surpassed—one of China's most sophisticated AI systems: DeepSeek.

OpenThinker-32B, developed by the Open Thoughts consortium, achieved a 90.6% accuracy score on the MATH500 benchmark, edging past DeepSeek's 89.4%.

The model also outperformed DeepSeek on general problem-solving tasks, scoring 61.6 on the GPQA-Diamond benchmark compared to DeepSeek's 57.6. On the LCBv2 benchmark, it hit a solid 68.9, showing strong performance across diverse testing scenarios.

...

all 9 comments

sorted by: hot top controversial new old

[–] HubertManne@moist.catsweat.com 6 points 2 days ago

I really feel a smaller high quality data set would beat feeding it anything and everything.

[–] autonomoususer@lemmy.world 11 points 4 days ago (1 children)

Is this libre software?

[–] Hotznplotzn@lemmy.sdf.org 16 points 4 days ago (2 children)

Model weights, datasets, data generation code, evaluation code, and training code are all publicly available.

[–] double_quack@lemm.ee 5 points 4 days ago (2 children)

Hey, I came late to the party. I am CS but I am far from AI. Can you point me to a resource where I can learn how to use all of "those things" that you mention?

[–] naeap@sopuli.xyz 9 points 4 days ago (1 children)

Had the same problem and someone guided me to the hugging face documents/tutorials.
They are quite nice to get a local model up and running, play around with it, how to fine tune it and connect it with agents

Haven't tried much, but the articles were exactly what I was looking for
Hope it helps you as well

[–] double_quack@lemm.ee 5 points 4 days ago

Thank you!

[–] Hotznplotzn@lemmy.sdf.org 6 points 4 days ago* (last edited 4 days ago)

As @naeap@sopuli.xyz said, it's on their Hugging Face site (here the link again: https://huggingface.co/open-thoughts/OpenThinker-32B), just below the first table are all the links.

[–] autonomoususer@lemmy.world 2 points 4 days ago* (last edited 4 days ago) (1 children)

But is it libre? Looks like it is all Apache 2.0, so yes.