We’re excited to introduce CSM-1, our Cyber Safety Model, built with a focus on understanding cyber threats, identifying risks, and strengthening AI-powered cyber safety.
CSM-1 represents an important step forward in our work towards smarter, more capable cyber safety systems.
And we’re not stopping here.
👀 CSM-1.1 is coming soon.
A new iteration, further improvements, and more to come.
Its our latest, ternary-weight model based on LFM 3 1.7B. Our target is 80-85% top 1 alignment, putting it close to 4 bit.
Klein Ternary is aimed at coding, with a footprint under 1 GB. It’s best suited for Raspberry Pis, phones and other low-power devices where even 4-bit models are too heavy.
The training pipeline is, normal self distillation, with a small seperate stage to bring out more coding performance after, and then a bit of post-training on Qwen distilled data to make it better.
We're announcing Klein SL Pro, our new model at Klein ML. It's doing post-training of Qwen3.8 Flash Next to make IT fit on 16GB VRAM.
It works by converting its MoE to SLMoE, SLMoE is an architecture, that chooses experts per response, not per token anymore.
This makes it so you can offload everything else to the disk and not get slow speed.
We're also seeking more compute to do this same thing on Deepseek V4.1 Flash, allowing it to run on 16GB VRAM tightly and definitely on 24GB. If you have hardware, more than 200B of VRAM in total, create a discussion klein-ml/README.
We've did some updates to Banaxi-Tech/TTS-Arena and now, everyone who uses the TTS Arena alot will get featured in our Klein TTS Release. Using it REALLY helps us alot so please use it.
Updates on the sizes: Klein TTS Flash Lite with 0.4M parameters, Klein TTS Flash with 1.7M parameters, Klein TTS Pro with 13M parameters and Ultra with 21M.
1.76× faster decoding on BF16, 1.65× on FP8, 1.51× on NVFP4. Agentic coding ~2.6×. Same outputs: exact rejection sampling. This is an early result, an improved version with higher acceptance is in training.
BananaMind 3 is going to be an comeback. Sorry for the BGA things ( i genuinely didnt know about nsa existing and yes I used aiiii) Were going to use our BananaMind 2.1 Unified architecture with some changes + NSA and some other stuff.