charmander_cha

Microsoft Aion 1.0 Instruct and Aion 1.0 Plan models!

Posted by Mysterious_Finish543@reddit | LocalLLaMA | View on Reddit | 109 comments

What's the status of non-CUDA inference?

Posted by IngwiePhoenix@reddit | LocalLLaMA | View on Reddit | 30 comments

charmander_cha@reddit

Texto eu uso vulkan, comfyui eu não tive problemas com rocm utilizando uma RX 7600 XT Já tts eu apenas não utilizo os o modelos mais novos pois não há suporte a minha língua nativa

StepFun 3.7 Flash

Posted by Everlier@reddit | LocalLLaMA | View on Reddit | 151 comments

Looks like Miminax-M3 is just around the corner

Posted by OnkelBB@reddit | LocalLLaMA | View on Reddit | 40 comments

[NEW] Supra-50M Released!

Posted by Dangerous_Try3619@reddit | LocalLLaMA | View on Reddit | 60 comments

Waiting on Qwen to drop those 3.7 models be like:

Posted by Porespellar@reddit | LocalLLaMA | View on Reddit | 47 comments

Carbon: Decoding the Language of Life

Posted by loubnabnl@reddit | LocalLLaMA | View on Reddit | 49 comments

charmander_cha@reddit

Soube deste modelo estes dias, eu não entendo desta tecnologia, só queria saber se está tecnologia se envolve de alguma forma com a sua tecnologia, não sobre o que elas fazem mas sobre como cada abordagem entende uma LLM e o que ela de fato faz com representações textuais de informação

Rewrite Bun in Rust has been merged

Posted by gruenistblau@reddit | programming | View on Reddit | 412 comments

GrapheneOS: Google's Play Integrity API requires hardware attestation ... Apple already has it as a requirement. Over the long term, this will increasingly lock out hardware and OS competition.

Posted by TheTwelveYearOld@reddit | linux | View on Reddit | 245 comments

Any news (or hope) of Qwen-3.6 14B and 9B distills for local coding ?

Posted by QuchchenEbrithin2day@reddit | LocalLLaMA | View on Reddit | 54 comments

Any news (or hope) of Qwen-3.6 14B and 9B distills for local coding ?

Posted by QuchchenEbrithin2day@reddit | LocalLLaMA | View on Reddit | 54 comments

You can now read Gemma 3's mind

Posted by DigiDecode_@reddit | LocalLLaMA | View on Reddit | 20 comments

charmander_cha@reddit

Esta pesquisa é daora, problema é que eu queria pesquisar os modelos chineses, pouco importa os americanos, além de poucos, são sempre os pequenos, se tiver uma forma de estudar os modelos de 9 ou 27B localmente seria incrível. Provavelmente isso foi lançado para "competir" em visibilidade com qwen scope

HOT TAKE: local models + agent harnesses are now capable enough to hand off junior-level IT professional tasks to [human written]

Posted by Porespellar@reddit | LocalLLaMA | View on Reddit | 67 comments

charmander_cha@reddit

Não sabia que isso era uma opinião polêmica, mas não é isso que vai matar os Junior, eles já estão mortos porque a tendência do capitalismo é monopólio, se é monopólio não tem como ter emprego para todos.

Introducing SubQ: The First Fully Subquadratic LLM

Posted by hltt@reddit | LocalLLaMA | View on Reddit | 6 comments

Bun is being rewritten to Rust

Posted by aabbdev@reddit | programming | View on Reddit | 152 comments

Open source models are going to be the future on Cursor, OpenCode etc.

Posted by _maverick98@reddit | LocalLLaMA | View on Reddit | 148 comments

Qwen 3.6 seems to have a lot of trouble with tool calling

Posted by Perfect-Campaign9551@reddit | LocalLLaMA | View on Reddit | 58 comments

Qwen Models are such good models?

Posted by FeiX7@reddit | LocalLLaMA | View on Reddit | 27 comments

Meta’s $2 billion Manus acquisition blocked by China.

Posted by Nunki08@reddit | LocalLLaMA | View on Reddit | 97 comments

charmander_cha@reddit

Literalmente as big techs estão envolvidas em genocídios. De um lado nos temos etnocentrismo, comum da humanidade. Do outro, uma big tech que tem levado a massacres e quedas de governos com intuito deixar a população em situação pior socialmente falando. Talvez seja a hora de você pesquisar sobre os crimes da Meta na África. E avaliar se seu comparativo ante de tudo, tem ou não sentido.

Meta’s $2 billion Manus acquisition blocked by China.

Posted by Nunki08@reddit | LocalLLaMA | View on Reddit | 97 comments

Opencode-power-pack – Claude Code skills ported to OpenCode

Posted by waybarrios@reddit | LocalLLaMA | View on Reddit | 8 comments

charmander_cha@reddit

Está na hora de começarmos a investirmos dinheiro nas APIs corretas para de forma organizada agirmos contra as big tech e implementar esse meio mundo de paper. Será que a comunidade conseguiria padronizar um método eficiente de implementar um paper? (na medida do possível)

Qwen 3.6 27B Makes Huge Gains in Agency on Artificial Analysis - Ties with Sonnet 4.6

Posted by dionysio211@reddit | LocalLLaMA | View on Reddit | 177 comments

Qwen3 TTS is seriously underrated - I got it running locally in real-time and it's one of the most expressive open TTS models I've tried

Posted by fagenorn@reddit | LocalLLaMA | View on Reddit | 117 comments

US gov memo on “adversarial distillation” - are we heading toward tighter controls on open models?

Posted by MLExpert000@reddit | LocalLLaMA | View on Reddit | 399 comments

Qwen-3.6-27B, llamacpp, speculative decoding - appreciation post

Posted by Then-Topic8766@reddit | LocalLLaMA | View on Reddit | 95 comments

Qwen3 TTS is seriously underrated - I got it running locally in real-time and it's one of the most expressive open TTS models I've tried

Posted by fagenorn@reddit | LocalLLaMA | View on Reddit | 117 comments

I don’t believe this benchmark 27b size model next opus 4.5! Anyone can confirm testing with real agentic workflow?

Posted by Wonderful-Ad-5952@reddit | LocalLLaMA | View on Reddit | 50 comments

MIT & the IMO released MathNet, the world’s largest dataset of International Math Olympiad problems & solutions. MathNet is 5x larger than previous datasets & is sourced from over 40 countries across 4 decades

Posted by Nunki08@reddit | LocalLLaMA | View on Reddit | 5 comments

Qwen 3.6 27B is out

Posted by NoConcert8847@reddit | LocalLLaMA | View on Reddit | 609 comments

llama.cpp is the linux of llm

Posted by DevelopmentBorn3978@reddit | LocalLLaMA | View on Reddit | 96 comments

Why doesn't any OSS tool treat llama.cpp as a first class citizen?

Posted by rm-rf-rm@reddit | LocalLLaMA | View on Reddit | 120 comments

charmander_cha@reddit

Se você fizer isso você vai fortalecer a narrativa do opensource/software livre, e eles defendem aqueles que funcionam mais próximos de serem produtos, com prospecção de lucros. É questão comercial, que exige de nós um posicionamento ético e também de propaganda

When did Github stop being about Git?

Posted by dgkimpton@reddit | programming | View on Reddit | 134 comments

Qwen3.6 (35B-A3B) with OpenCode. Running locally with llama.cpp

Posted by curiousily_@reddit | LocalLLaMA | View on Reddit | 5 comments

Are you guys actually using local tool calling or is it a collective prank?

Posted by Mayion@reddit | LocalLLaMA | View on Reddit | 197 comments

When is Qwen 3.6 27B dropping? Didn’t it win the vote?

Posted by GrungeWerX@reddit | LocalLLaMA | View on Reddit | 72 comments

Don't ask Qwen 3.6 35b to give you aski image of Yoshi :)

Posted by anzzax@reddit | LocalLLaMA | View on Reddit | 19 comments

Qwen3.6. This is it.

Posted by Local-Cardiologist-5@reddit | LocalLLaMA | View on Reddit | 420 comments

Bonsai models are pure hype: Bonsai-8B is MUCH dumber than Gemma-4-E2B

Posted by WeGoToMars7@reddit | LocalLLaMA | View on Reddit | 71 comments

Ternary Bonsai: Top intelligence at 1.58 bits

Posted by pmttyji@reddit | LocalLLaMA | View on Reddit | 89 comments

Ternary Bonsai: Top intelligence at 1.58 bits

Posted by pmttyji@reddit | LocalLLaMA | View on Reddit | 89 comments

Ternary Bonsai: Top intelligence at 1.58 bits

Posted by pmttyji@reddit | LocalLLaMA | View on Reddit | 89 comments

Comparison Qwen 3.6 35B MoE vs Qwen 3.5 35B MoE on Research Paper to WebApp

Posted by dreamai87@reddit | LocalLLaMA | View on Reddit | 29 comments

Its just a new Qwen model

Posted by StandardLovers@reddit | LocalLLaMA | View on Reddit | 21 comments

VAD issues - takes too much time to understand when the user has stopped talking

Posted by Male_Cat_@reddit | LocalLLaMA | View on Reddit | 9 comments

(llama.cpp) Possible to disable reasoning for some requests (while leaving reasoning on by default)?

Posted by regunakyle@reddit | LocalLLaMA | View on Reddit | 21 comments

(llama.cpp) Possible to disable reasoning for some requests (while leaving reasoning on by default)?

Posted by regunakyle@reddit | LocalLLaMA | View on Reddit | 21 comments

BlueTTS is basically supertonic look at the paper and the code

Posted by Elegant-Condition206@reddit | LocalLLaMA | View on Reddit | 6 comments

huge improvement after moving from ollama to llama.cpp

Posted by leonardosalvatore@reddit | LocalLLaMA | View on Reddit | 76 comments

It was fun while it lasted...

Posted by Specter_Origin@reddit | LocalLLaMA | View on Reddit | 15 comments

It was fun while it lasted...

Posted by Specter_Origin@reddit | LocalLLaMA | View on Reddit | 15 comments

charmander_cha@reddit

E porque a porra dos Estados Unidos bombardeou uma nação que literalmente estava quieta (ao contrário do que diz a propaganda naziamericana) e agora porque esses filhos da puta possuem como hobby matar crianças e principalmente garotas, teremos que pagar a conta do petróleo. Nação governada por pedófilos que governam burros racistas