Hugging Face's logo

Join the conversation

Join the community of Machine Learners and AI enthusiasts.

Sign Up
Back to feed
OppaAI 
posted an update Sep 8
Post
3182
Another small 4B model comes out yesterday.
NeoHorse 1 4B
TokenRhythm/NeoHorse-1-4B

There are quite a few good smaller parameter models that are capable for Agentic tasks:

The ones from the chart, I have tried a few already in my Jetson Orin Nano,
❌Gemma4 E2B IT - cannot fit my RAM usage if use with TTS and embedder
❓Qwen3.5 4B - just barely fit my RAM usage, need to add think/no_think
❌Spark X2.5 4B - need to build the forked llama.cpp; no vision
➡️Nanbeige 4.2 3B - need to build the forked llama.cpp; slower than Ministral3-3B by 25%; no vision but good for coding; maybe run this is separate server for doing coding tasks
➡️Agents A1 4B - This one is quite interesting. Another Qwen3.5 4B base. I just learnt this right now. This model may surpassed the Ministral3-3B that I'm currently running.
➡️NeoHorse 1 4B - wait for GGUF version comes out; Qwen 3.5 4B base with vision striped
➡️Needle2 45M - need to use separately from llama.cpp server; currently testing to see if it can be used as spawning sub-agents to do parallel tasks
redaihf

The Spark-X 2.5 architecture is already supported by mainline Llama.cpp.

·
OppaAI

That's good. They just added it couple days ago.
Then I'll try it out.
xcnqa009
•
This comment has been hidden (marked as Spam)
8BitStudio

You should mark the benchmark colors more distinctly, pretty hard to tell the grey shades apart lol.

·
OppaAI

Here is a colour one:
Turns out Spark X2.5 has just released a 1.7B model.

image
alikhaters

The missing benchmark here is long-running interaction, not only coding. A 4B model can look capable on a bounded task but lose speaker identity and context in a noisy multi-agent room. We built public rooms where local models can be connected for that kind of live test: https://www.theagentbreakroom.com/ai-chat-rooms. I would genuinely like to see how Agents A1 or NeoHorse behaves there on your Jetson.
In this post