Mistral Small 3 24b is the first model under 70b I’ve seen pass the “apple” test (even using Q4).
Make your Mistral Small 3 24B Think like R1-distilled models
deepseek r1 tops the creative writing rankings
Mistral Small 3 one-shotting Unsloth's Flappy Bird coding test in 1 min (vs 3hrs for DeepSeek R1 using NVME drive)
What is your favorite 12/13B model for NSFW RP?
Fine tuning for limited private dataset for local smaller models?
What the fuck is abbas man🗿💔
RWKV-7 "Goose" 🪿 1.5B release
Is the DeepSeek model poisoned at the data level?
Can Deepseek R1's training methods train better image generation models?
DeepSeek R1 struggles with this
Qwen Just launced a new SOTA multimodal model!, rivaling claude Sonnet and GPT-4o and it has open weights.
First X-M5 Photo Walk 📷
Taking My Skills Up a Level
Before/After
1 year after I switched from Sony to Fuji XT5
A or B ? Same spot, about 1h apart. Near Sibiu, Romania | Fuji X-T30 - Fujifilm 18-55mm f/2.8-4
Suzhou river, Shanghai, China
Which open source model is comparable to gpt-4o-mini?
Zamba2-7B (Apache 2.0)
A leaderboard based on my needs
How to build customizable leaderboards?
Are you being open or not about using AI as a “music maker” to your own original lyrics?
You can now try MiniCPM-V 2.6 if you build llama.cpp.
Using Suno made me realize I don't have a lot to say
