Agnes-AI/Agnes-3.0-Qwen
Image-Text-to-Text • 180B • Updated • 87 • 6
Rev Quick
Revertron
AI & ML interests
None yet
Recent Activity
liked a model 1 day ago
Agnes-AI/Agnes-3.0-Qwen liked a model 2 days ago
KittenML/kitten-tts-2 new activity 5 days ago
IFM/K2-Horizon-7B:I wish you delivered what you say you deliveredOrganizations
None yet
liked a
model 1 day ago
liked a
model 2 days ago
I wish you delivered what you say you delivered
👍👀 4
15
#3 opened about 1 month ago
by 
NezTheNaughty
reacted to mihailgribov's post with 🤯 6 days ago
Post
2307
TypeCastLM: Jev-class decision models on frozen LLMs
TypeCastLM is a decision model built around a frozen LLM. Given a text and a question, it returns calibrated probabilities over the allowed answers in one forward pass. The LLM is not fine-tuned: not one of its weights changes. TypeCastLM swaps its LM head for a mini one: a linear matrix whose rows are mostly the model's own output rows, with the three verdict rows (true, false, unsure) fitted. A fitted row catches an answer the model spreads over many tokens, such as "yes", "true" or "correct", where a single vocabulary row sees one word. A head that small leaves little room to overfit. It speaks the Jev API, so a Jev client switches by changing the base URL.
The first model, [typecastlm-qwen3.5-3.8b]( mihailgribov/typecastlm-qwen3.5-3.8b) on Qwen3.5-4B, is among the top open frozen-4B models on JevBench v1.6. Besides Jev's yes/no, choice and scale modes it has
Specs:
• Size: 3.76B params; 7.5 GB bf16, 4.0 GB Q8_0
• VRAM: 12 GB is enough, 16 GB comfortable
• Latency p50 (batch 1, bf16, RTX 5060 Ti): ≤200 tok 48 ms · 1k 93 ms · 4k 566 ms
• Calibration (ECE): BoolQ 0.052 · RTE 0.012 · FEVER 0.019
• Context: 32k tokens
• Runtime: transformers or llama.cpp, fully offline
• JevBench v1.6: 19.7
TypeCastLM is a decision model built around a frozen LLM. Given a text and a question, it returns calibrated probabilities over the allowed answers in one forward pass. The LLM is not fine-tuned: not one of its weights changes. TypeCastLM swaps its LM head for a mini one: a linear matrix whose rows are mostly the model's own output rows, with the three verdict rows (true, false, unsure) fitted. A fitted row catches an answer the model spreads over many tokens, such as "yes", "true" or "correct", where a single vocabulary row sees one word. A head that small leaves little room to overfit. It speaks the Jev API, so a Jev client switches by changing the base URL.
The first model, [typecastlm-qwen3.5-3.8b]( mihailgribov/typecastlm-qwen3.5-3.8b) on Qwen3.5-4B, is among the top open frozen-4B models on JevBench v1.6. Besides Jev's yes/no, choice and scale modes it has
tfu: yes/no with a third answer, unsure, for when the text lacks what the decision needs.Specs:
• Size: 3.76B params; 7.5 GB bf16, 4.0 GB Q8_0
• VRAM: 12 GB is enough, 16 GB comfortable
• Latency p50 (batch 1, bf16, RTX 5060 Ti): ≤200 tok 48 ms · 1k 93 ms · 4k 566 ms
• Calibration (ECE): BoolQ 0.052 · RTE 0.012 · FEVER 0.019
• Context: 32k tokens
• Runtime: transformers or llama.cpp, fully offline
• JevBench v1.6: 19.7
pip install typecastlm 
- 2 replies
liked a
model 7 days ago
liked 2
models 8 days ago
liked a
dataset 9 days ago
Great goal! I'm also trying something like that, but more as an SFT for fine-tuning. But now I've pivoted to a sophisticated library inside my coding-agent...
reacted to cafkafk's post with 🔥 12 days ago
Post
4490
I'm working on a local model that can be run on stuff like an an RTX 3080 at 60 t/s. I'm focused on making it good at Rust and Nix. It's... not done yet, but doing great already. Had a 28/100 on livebench v6 earlier, which obviously isn't amazing.
But for my first real model, very exciting. Can't wait to share more as I get closer to the finish line in the coming weeks.
But for my first real model, very exciting. Can't wait to share more as I get closer to the finish line in the coming weeks.
- 6 replies
liked a
dataset 13 days ago
liked 2
models 14 days ago
liked a
dataset 15 days ago
It can't use tools?
#1 opened 15 days ago
by Revertron
liked a
model 19 days ago
liked a
dataset 20 days ago
liked a
model 21 days ago
liked a
model 23 days ago
