Banaxi-Tech/bananamind-arena-logs
837 kB
Banaxi PRO
Banaxi-Tech
AI & ML interests
SLMs, training from scratch, LoRA, TTS, Ternary models. AI Interpretability. BCI. Contact at banaxitech@gmail.com
Recent Activity
repliedto their post about 1 hour ago
We're announcing KV1. Now.
We're going to take some models, do a bit of continued training on them and QAT them for only a 2 bit KV cache.
And because of what happened with BGA, and this is also kind of some research stuff, here's what we're going to change:
We're going to always, before we release how and make any claims, release our results.
We also would like (idk if we can) to have some external reviewers before releasing them, I'M talking to YOU @AtAndDev and @mithulaartigala
And we're going to properly cite other research.
The models we are going to release is:
KV1 1.2B, based on LFM 2.5 1.2B Instruct
KV1 2.6B, based on LFM 2.5 2.6B Instruct
KV1 4B, based on Qwen3.5 4B.
If we get enough compute, we may also train:
KV1 9B
And if you guys want to a ~27B model, you guys decide which, either on LFM 2.5 24B A2B, Qwen3.8 27B, or Qwen3.6 35B A3B.
Our goal is to get 2 bit, at 4 bit quality.
Here's an image of our current progress, KV1 is at 3 bit, not 4 bit, but it will improve! repliedto their post about 1 hour ago
We're announcing KV1. Now.
We're going to take some models, do a bit of continued training on them and QAT them for only a 2 bit KV cache.
And because of what happened with BGA, and this is also kind of some research stuff, here's what we're going to change:
We're going to always, before we release how and make any claims, release our results.
We also would like (idk if we can) to have some external reviewers before releasing them, I'M talking to YOU @AtAndDev and @mithulaartigala
And we're going to properly cite other research.
The models we are going to release is:
KV1 1.2B, based on LFM 2.5 1.2B Instruct
KV1 2.6B, based on LFM 2.5 2.6B Instruct
KV1 4B, based on Qwen3.5 4B.
If we get enough compute, we may also train:
KV1 9B
And if you guys want to a ~27B model, you guys decide which, either on LFM 2.5 24B A2B, Qwen3.8 27B, or Qwen3.6 35B A3B.
Our goal is to get 2 bit, at 4 bit quality.
Here's an image of our current progress, KV1 is at 3 bit, not 4 bit, but it will improve! repliedto their post about 2 hours ago
We're announcing KV1. Now.
We're going to take some models, do a bit of continued training on them and QAT them for only a 2 bit KV cache.
And because of what happened with BGA, and this is also kind of some research stuff, here's what we're going to change:
We're going to always, before we release how and make any claims, release our results.
We also would like (idk if we can) to have some external reviewers before releasing them, I'M talking to YOU @AtAndDev and @mithulaartigala
And we're going to properly cite other research.
The models we are going to release is:
KV1 1.2B, based on LFM 2.5 1.2B Instruct
KV1 2.6B, based on LFM 2.5 2.6B Instruct
KV1 4B, based on Qwen3.5 4B.
If we get enough compute, we may also train:
KV1 9B
And if you guys want to a ~27B model, you guys decide which, either on LFM 2.5 24B A2B, Qwen3.8 27B, or Qwen3.6 35B A3B.
Our goal is to get 2 bit, at 4 bit quality.
Here's an image of our current progress, KV1 is at 3 bit, not 4 bit, but it will improve!Organizations
Posts 69
Post
414
KV1 updates:
We at klein-ml have finished stage 1 of KV1 1.2B, we don't think we will need additional stages.
KV1 2.6B is near the midpoint of training, going well so far, its going to be our most agentic model yet.
We have released Klein Code Bench klein-ml/klein-code-bench, a benchmark testing whether models can build complete projects from scratch. It contains a easy, intermediate, hard and max subset.
Here are some benchmarks of the current KV1 1.2B
It retains 75.8% of FP16 intelligence.
We are also announcing that we at klein-ml are working on Klein 24B A2B which will be our first model featuring ternary weight quantization.
We at
KV1 2.6B is near the midpoint of training, going well so far, its going to be our most agentic model yet.
We have released Klein Code Bench klein-ml/klein-code-bench, a benchmark testing whether models can build complete projects from scratch. It contains a easy, intermediate, hard and max subset.
Here are some benchmarks of the current KV1 1.2B
It retains 75.8% of FP16 intelligence.
We are also announcing that we at
spaces 10
pinned
Running on Zero
Agents
6
BananaMind Arena
🍌
Blind Elo battles between very small language models
Running on Zero
Agents
3
KV1 Chat
💬
Chat with KV1 AI model for instant replies
Running
1
Release Radar
🍌
View upcoming model release dates with probability forecasts
Running on Zero
Agents
6
BananaMind 2 Pro Preview Demo
🍌
Chat with BananaMind 2 Pro AI model
Running on Zero
Agents
1
BananaMind 2 Medium Demo
👀
Chat with BananaMind 2 Medium language model
Running on Zero
Agents
3
BananaMind 2 AI Detect
🍌
Human-vs-AI text detector (Qwen3.5-0.8B fine-tuned on RAID)
models 38
Banaxi-Tech/test-model
Updated
Banaxi-Tech/face-model
Object Detection • Updated • 58
Banaxi-Tech/mess-around-with-bannamindbot
Updated
Banaxi-Tech/BananaMind-2.1-Lite-All-Looped-AdamW
Text Generation • 28.4M • Updated • 572 • 3
Banaxi-Tech/BananaMind-2.1-Lite-AdamW
Text Generation • 28.4M • Updated • 152 • 1
Banaxi-Tech/BananaMind-2.1-Flash-Lite-AdamW
12.3M • Updated • 53
Banaxi-Tech/BananaMind-2.1-Lite
Text Generation • 28.4M • Updated • 161 • 2
Banaxi-Tech/BananaMind-2.1-Flash-Lite
12.3M • Updated • 45
Banaxi-Tech/BananaMind-2.1-NanoCoder
Text Generation • 12.3M • Updated • 111 • 1
Banaxi-Tech/BananaMind-2.1-MiniCoder
Text Generation • 28.4M • Updated • 139 • 2