Didn't realize R1 Distilled Qwen 32B was THIS good!
How much monster do you think is too much in one day?
I trained a tinystories model from scratch for educational purposes, how cooked? (1M-parameters)
Tried sour ropes!
Macbook Air 15 for AI ?
The new Mistral Small model is disappointing
Best model for book translation
I want to roast some AI projects today 🤨. If you are not afraid, share your project in the comments!
Mistral Small 3 24b Q6 initial test results
QWEN just launched their chatbot website
Mistral Small 3
Which models don't have that "AI" look
Real news: 32B distills of V3, soon R1.
DeepSeek R1 just got a 2x speed boost. The crazy part? The code for the boost was written by R1 itself. Self-improving AI is here.
Ollama is confusing people by pretending that the little distillation models are "R1"
Just canceled my OpenAI Plus subscription (for now). Been running DeepSeek-R1 14b locally on my home workstation. I'll probably renew it if OpenAI launches something worthy for Plus tier by then.
Jealous?
The same prompt in Janus-Pro-7B, Dall-e and Flux Dev
freePreTrainedModelIsCoolThough
Qwen2.5 14B-1M Impressions
UPDATE: Monster Creme Brûlée
Your caffeine is ready sir
deepseek r1 tops the creative writing rankings
Censorship has started to get worse.
What's the most influential model since BERT?
