janhq/Jan-v3.5-4B
Text Generation • 4B • Updated • 219 • 11
Jan Team
jan-ai
AI & ML interests
Also at https://huggingface.co/janhq
Recent Activity
published a model 16 days ago
janhq/Jan-v3.5-4B published a model 16 days ago
janhq/Jan-v3.5-4B-gguf updated a model 16 days ago
janhq/Jan-v3.5-4BOrganizations
published 2
models 16 days ago
updated 2
models 16 days ago
updated a
collection 27 days ago
updated a
model 28 days ago
updated 2
models about 1 month ago
published 2
models about 1 month ago
reactedto alandao's post with 🔥 about 1 month ago
Post
1466
Don’t give up 🔥
Do you know what I was planning to do this time last week?
I was preparing to write a report declaring that Jan Nano was a failed project because the benchmark results didn’t meet expectations.
But I thought — it can’t be. When loading the model into the app, the performance clearly felt better. So why were the benchmark results worse?
That’s when I reviewed the entire benchmark codebase and realized something fundamental: agentic or workflow-based approaches introduce a huge gap and variation when benchmarking. Jan-nano was trained with an agentic setup — it simply can’t be benchmarked using a rigid workflow-based method.
I made the necessary changes, and the model ended up performing even better than before the issues arose. Turns out the previous benchmarking method conflicted with the way the model was trained.
What if I had given up? That would’ve meant 1.5 months of training and a huge amount of company resources wasted.
But now, this is officially the most successful and biggest release for the whole team — all thanks to Jan-nano.
Menlo/Jan-nano
Do you know what I was planning to do this time last week?
I was preparing to write a report declaring that Jan Nano was a failed project because the benchmark results didn’t meet expectations.
But I thought — it can’t be. When loading the model into the app, the performance clearly felt better. So why were the benchmark results worse?
That’s when I reviewed the entire benchmark codebase and realized something fundamental: agentic or workflow-based approaches introduce a huge gap and variation when benchmarking. Jan-nano was trained with an agentic setup — it simply can’t be benchmarked using a rigid workflow-based method.
I made the necessary changes, and the model ended up performing even better than before the issues arose. Turns out the previous benchmarking method conflicted with the way the model was trained.
What if I had given up? That would’ve meant 1.5 months of training and a huge amount of company resources wasted.
But now, this is officially the most successful and biggest release for the whole team — all thanks to Jan-nano.
Menlo/Jan-nano
updated 2
models about 2 months ago
Update README.md
#3 opened 2 months ago
by 
jan-ai
liked 2
models 2 months ago
updated a
model 10 months ago
published a
model 10 months ago
updated a
model 10 months ago
published a
model 10 months ago