Qwen3-VL
Collection
37 items • Updated • 781
🏗️ Building on HF
A. Batuhan Bilgin
abatuhanb
AI & ML interests
Natural Language Processing (NLP)
AI Researcher & HPC
Recent Activity
upvoted a collection 5 days ago
Qwen3-VL liked a model 20 days ago
Qwen/Qwen2.5-VL-32B-InstructOrganizations
upvoted a collection 5 days ago
reacted to onekq's post with 👍 6 days ago
Post
3176
My takes on Jev
1. Very likely a small model. You can certainly pretrain, but I would grab an existing base model, say Qwen 3 class
2. The new RL method is a breakthrough, classification doesn't need to align with human preferences
3. The new output is an overstatement. It's just a new LM head. Of course autoregressive decoding can be used for classification: it takes just a few tokens to express the output. Think twice: are you sure classification doesn't need few-shot, CoT, or reasoning? All of these depend on auto-regressiveness
4. It carves out a market already existing, which is now served by oversized LLMs (hence overpaid), e.g. LLM as judge, labeling
5. Jevons effect will kick in, promoting more modeling efforts for small budget teams. It might even accelerate RSI
1. Very likely a small model. You can certainly pretrain, but I would grab an existing base model, say Qwen 3 class
2. The new RL method is a breakthrough, classification doesn't need to align with human preferences
3. The new output is an overstatement. It's just a new LM head. Of course autoregressive decoding can be used for classification: it takes just a few tokens to express the output. Think twice: are you sure classification doesn't need few-shot, CoT, or reasoning? All of these depend on auto-regressiveness
4. It carves out a market already existing, which is now served by oversized LLMs (hence overpaid), e.g. LLM as judge, labeling
5. Jevons effect will kick in, promoting more modeling efforts for small budget teams. It might even accelerate RSI
- 7 replies
liked a
model 20 days ago
liked a
model 25 days ago
reacted to Banaxi-Tech's post with ❤️ 3 months ago
Post
10768
A new model is coming!
Its going to take a long time on my 5070 Ti so expect a release in ~1 month.
We think this model is going to be SOTA For its size.
Our Mini Version will be 25M Parameters and Pro with 140M.
The Pro version has a 3072 Context Window (Extensible to up to 6K with RoPE) And the Mini version has a context window of 4096 (Up to 8K with RoPE)
Meanwhile we are currently working on a Instruct Version of our BananaMind 1.5 Base.
The training will start this weekend
We are very exited to release it when its done!
Its going to take a long time on my 5070 Ti so expect a release in ~1 month.
We think this model is going to be SOTA For its size.
Our Mini Version will be 25M Parameters and Pro with 140M.
The Pro version has a 3072 Context Window (Extensible to up to 6K with RoPE) And the Mini version has a context window of 4096 (Up to 8K with RoPE)
Meanwhile we are currently working on a Instruct Version of our BananaMind 1.5 Base.
The training will start this weekend
We are very exited to release it when its done!
- 12 replies
liked 2
datasets 3 months ago
upvoted an article 3 months ago
Article
Building autograd engine tinytorch 01
joey00072
• • 5 liked a
model 4 months ago
liked a
model 6 months ago
upvoted an article 6 months ago
Article
Welcome Gemma 4: Frontier multimodal intelligence on device
- +5
merve, pcuenq, sergiopaniego, burtenshaw, Steveeeeeeen, alvarobartt, SaylorTwift
• • 929 liked 2
models 6 months ago
liked a
model 7 months ago
upvoted a collection 7 months ago
liked a
dataset 7 months ago
liked a
model 7 months ago
The model performs poorly.
1
#14 opened 8 months ago
by 
wyb5210
liked a
model 7 months ago
Regarding Instruction Following SFT dataset.
1
#22 opened about 1 year ago
by 
doubleyyh