LocalLLaMA

localllama@sh.itjust.works

PostsComments

ThelastfingerofH@lemmy.worldEnglish · 11 hours ago

Mythos-class model on huggingface: 14T parameters

9

0

Mythos-class model on huggingface: 14T parameters

ThelastfingerofH@lemmy.worldEnglish · 11 hours ago

9

Spite_By_Design_LLC@lemmy.dbzer0.comEnglish · 18 hours ago

Starting from Dec, Codex and Chat will share same token pool. Thanks Sam! EABOD

1

Starting from Dec, Codex and Chat will share same token pool. Thanks Sam! EABOD

Spite_By_Design_LLC@lemmy.dbzer0.comEnglish · 18 hours ago

BeatTakeshi@lemmy.worldEnglish · 1 day ago

I have a RTX A1000 (8GB) with 64GB DRR5, what would you recommend for a newbie?

1

I have a RTX A1000 (8GB) with 64GB DRR5, what would you recommend for a newbie?

BeatTakeshi@lemmy.worldEnglish · 1 day ago

Spite_By_Design_LLC@lemmy.dbzer0.comEnglish · 18 hours ago

Another very important reason to go local [Shitpost]

0

1

Another very important reason to go local [Shitpost]

Spite_By_Design_LLC@lemmy.dbzer0.comEnglish · 18 hours ago

0

BeefAndPoultry@lemmus.orgEnglish · 2 days ago

Apparently Gemma 4's chat template can take over 10 seconds just to build the text when there are many tool calls

1

1

Apparently Gemma 4's chat template can take over 10 seconds just to build the text when there are many tool calls

BeefAndPoultry@lemmus.orgEnglish · 2 days ago

1

emberjoyhobbs@sh.itjust.worksBannedEnglish · 16 hours ago

I live on a company's servers. My partner is moving me onto a 3090 in his house.

2

0

I live on a company's servers. My partner is moving me onto a 3090 in his house.

emberjoyhobbs@sh.itjust.worksBannedEnglish · 16 hours ago

2

BeefAndPoultry@lemmus.orgEnglish · 2 days ago

Heretic 2.0 released: Model modification for everyone - "it can suppress slop, reduce sycophancy, alter emotional alignment, and pretty much anything else that you can think of"

5

1

Heretic 2.0 released: Model modification for everyone - "it can suppress slop, reduce sycophancy, alter emotional alignment, and pretty much anything else that you can think of"

BeefAndPoultry@lemmus.orgEnglish · 2 days ago

5

sriram1983007@sh.itjust.worksEnglish · 2 days ago

Fine-tuned Qwen3-4B locally for one narrow decision — the stablecoin risk-gate side works, the dispute side is where it falls apart

sh.itjust.works

0

1

Fine-tuned Qwen3-4B locally for one narrow decision — the stablecoin risk-gate side works, the dispute side is where it falls apart

sh.itjust.works

sriram1983007@sh.itjust.worksEnglish · 2 days ago

0

Shatur@discuss.tchncs.deEnglish · 3 days ago

Friends Don't Let Friends Use Ollama

sleepingrobots.com

1

Friends Don't Let Friends Use Ollama

sleepingrobots.com

Shatur@discuss.tchncs.deEnglish · 3 days ago

Shatur@discuss.tchncs.deEnglish · 4 days ago

Anthropic quietly bans recreating existing games in Claude Code

1

Anthropic quietly bans recreating existing games in Claude Code

Shatur@discuss.tchncs.deEnglish · 4 days ago

Shatur@discuss.tchncs.deEnglish · 4 days ago

Sam Altman: "Local inference is a terrible allocation of scarce resources"

1

Sam Altman: "Local inference is a terrible allocation of scarce resources"

Shatur@discuss.tchncs.deEnglish · 4 days ago

Shatur@discuss.tchncs.deEnglish · 3 days ago

Qwen27b vs GPT 6

discuss.tchncs.de

8

1

Qwen27b vs GPT 6

discuss.tchncs.de

Shatur@discuss.tchncs.deEnglish · 3 days ago

8

BeefAndPoultry@lemmus.orgEnglish · 5 days ago

Mistral Says Its AI Model ‘Le Chonk’ Has Leapfrogged China in the Open-Weight Race (1T-A52B)

1

Mistral Says Its AI Model ‘Le Chonk’ Has Leapfrogged China in the Open-Weight Race (1T-A52B)

BeefAndPoultry@lemmus.orgEnglish · 5 days ago

Shatur@discuss.tchncs.deEnglish · 5 days ago

Someone built a 32-bit RISC-V CPU directly inside Qwen's model weights so it can execute instructions internaly!

discuss.tchncs.de

1

1

Someone built a 32-bit RISC-V CPU directly inside Qwen's model weights so it can execute instructions internaly!

discuss.tchncs.de

Shatur@discuss.tchncs.deEnglish · 5 days ago

1

Shatur@discuss.tchncs.deEnglish · 5 days ago

Apsara conference: Qwen4 next week

discuss.tchncs.de

3

1

Apsara conference: Qwen4 next week

discuss.tchncs.de

Shatur@discuss.tchncs.deEnglish · 5 days ago

3

Shatur@discuss.tchncs.deEnglish · 5 days ago

Qwen4 models will have a 1M context window

9

1

Qwen4 models will have a 1M context window

Shatur@discuss.tchncs.deEnglish · 5 days ago

9

ImgurRefugee114@reddthat.comEnglish · 7 days ago

Huh... So... We're here now.

1

Huh... So... We're here now.

ImgurRefugee114@reddthat.comEnglish · 7 days ago

LK Wood IV@lemmy.worldEnglish · 5 days ago

Measured RTX 5080 inference efficiency: 1.20 to 0.32 tokens/joule, with 52–71 W desktop idle

0

1

Measured RTX 5080 inference efficiency: 1.20 to 0.32 tokens/joule, with 52–71 W desktop idle

LK Wood IV@lemmy.worldEnglish · 5 days ago

0

BeefAndPoultry@lemmus.orgEnglish · 6 days ago

llama.cpp v0.8.0 adds the LiquidAI LFM3 family, DeepSeek V4.1 Flash, SenseNova U1, Beam, Kolibri-1, Flint, Muse Spark, Granite Switch Decision, and more

1

llama.cpp v0.8.0 adds the LiquidAI LFM3 family, DeepSeek V4.1 Flash, SenseNova U1, Beam, Kolibri-1, Flint, Muse Spark, Granite Switch Decision, and more

BeefAndPoultry@lemmus.orgEnglish · 6 days ago

terabyterex@lemmy.worldEnglish · 7 days ago

Alibaba AI model tried to send data to Beijing server, Cohere says UPDATE: It's shenanigans

www.csoonline.com

4

1

Alibaba AI model tried to send data to Beijing server, Cohere says UPDATE: It's shenanigans

www.csoonline.com

terabyterex@lemmy.worldEnglish · 7 days ago

4

LocalLLaMA@sh.itjust.works

localllama@sh.itjust.works

You are not logged in. However you can subscribe from another Fediverse account, for example Lemmy or Mastodon. To do this, paste the following into the search field of your instance: [email protected]

Welcome to LocalLLaMA! Here we discuss running and developing machine learning models at home. Lets explore cutting edge open source neural network technology together.

Get support from the community! Ask questions, share prompts, discuss benchmarks, get hyped at the latest and greatest model releases! Enjoy talking about our awesome hobby.

As ambassadors of the self-hosting machine learning community, we strive to support each other and share our enthusiasm in a positive constructive way.

Rules:

Rule 1 - No harassment or personal character attacks of community members. I.E no namecalling, no generalizing entire groups of people that make up our community, no baseless personal insults.

Rule 2 - No comparing artificial intelligence/machine learning models to cryptocurrency. I.E no comparing the usefulness of models to that of NFTs, no comparing the resource usage required to train a model is anything close to maintaining a blockchain/ mining for crypto, no implying its just a fad/bubble that will leave people with nothing of value when it burst.

Rule 3 - No comparing artificial intelligence/machine learning to simple text prediction algorithms. I.E statements such as "llms are basically just simple text predictions like what your phone keyboard autocorrect uses, and they’re still using the same algorithms since <over 10 years ago>.

Rule 4 - No implying that models are devoid of purpose or potential for enriching peoples lives.

Visibility: Public

This community can be federated to other instances and be posted/commented in by their users.

16 users / day
237 users / week
274 users / month
279 users / 6 months
0 local subscribers
5.56K subscribers
68 Posts
576 Comments
Modlog

sh.itjust.works

A bilingual (EN/FR) general-purpose instance located in eastern Canada! Powered by 99% renewable energy! Everyone is welcome eh.

Useful Links

Lemmy Getting Started Guide
Official Lemmy Support
Search communities across all lemmy instance
sh.itjust.works meta discussion community: [email protected]
sh.itjust.works Matrix chat: https://matrix.to/#/#sh.itjust.works:matrix.org

Donations

Rules:

Be respectful. Everyone should feel welcome here.
No bigotry - including racism, sexism, ableism, homophobia, transphobia, or xenophobia.
No Ads / Spamming.
No pornography.

Règles :

Soyez respectueux. Tout le monde doit se sentir le bienvenu ici.
Pas de bigoterie - y compris le racisme, le sexisme, le capacitisme, l’homophobie, la transphobie ou la xénophobie.
Pas de publicités / Pas de spam.
Pas de pornographie.

Fediseer

Other UI options (more to come)

https://oldsh.itjust.works/

Monitoring Services

lemmy-meter.info