zai-org/GLM-5.3
Text Generation • 753B • Updated • 66.2k • • 1.41k
jungletsubasa
jungletsubasa
·
AI & ML interests
None yet
Recent Activity
liked a model 3 days ago
zai-org/GLM-5.3 new activity 7 days ago
lukealonso/Kimi-K3-QSRT-K2:Thank you for the model! liked a model 12 days ago
lukealonso/Kimi-K3-QSRT-K2Organizations
None yet
liked a
model 3 days ago
Thank you for the model!
#1 opened 13 days ago
by jungletsubasa
liked a
model 12 days ago
reacted to onekq's post with 🔥 13 days ago
Post
3228
Lots of attentions are now on GPU residual value. I find car analogy to be useful.
* Both new and used cars can do the same job (you can run the latest model on 6-year-old A100)
* New cars are more efficient (higher performance per power draw)
* It's not the year, but mileage and maintenance
The last point is undeveloped. We need Carfax and KBB for GPUs.
And there will be lemon GPUs.
* Both new and used cars can do the same job (you can run the latest model on 6-year-old A100)
* New cars are more efficient (higher performance per power draw)
* It's not the year, but mileage and maintenance
The last point is undeveloped. We need Carfax and KBB for GPUs.
And there will be lemon GPUs.
- 1 reply
liked 4
models 24 days ago
baseten/GLM-5.2-Vision-FP8
Image-Text-to-Text • 754B • Updated • 528 • 12
deepseek-ai/DeepSeek-V4-Pro-DSpark
Text Generation • 1.7T • Updated • 4.2k • 540
deepseek-ai/DeepSeek-V4-Flash-0731
Text Generation • 304B • Updated • 4.56M • • 3.84k
FlyCockpit/DeepSeek-V4-Flash-0731-vision
Image-Text-to-Text • Updated • 285 • 10
liked a
Space about 1 month ago
liked 2
models about 1 month ago
very very nice and smart
2
#1 opened 3 months ago
by jungletsubasa
Any llama.cpp parameters to work around the looping?
👍 1
2
#12 opened 4 months ago
by 
tarruda
IQuest-Coder-V1 Model Family Update 40B-Thinking
1
#1938 opened 6 months ago
by jungletsubasa
reacted to SeaWolf-AI's post with 👀🤗🤝 6 months ago
Post
4346
FINAL Bench Released: The Real Bottleneck to AGI Is Self-Correction
We release FINAL Bench, the first benchmark for measuring functional metacognition in LLMs — the ability to detect and correct one's own reasoning errors. Every existing benchmark measures final-answer accuracy. None measures whether AI knows it is wrong.
Dataset: [FINAL-Bench/Metacognitive]( FINAL-Bench/Metacognitive) | 100 Tasks | 15 Domains | 8 TICOS Types | Apache 2.0
Leaderboard: FINAL-Bench/Leaderboard
Article: https://huggingface.co/blog/FINAL-Bench/metacognitive
Core Innovation
Our 5-axis rubric separates what no prior benchmark could: MA (Metacognitive Accuracy) — the ability to say "I might be wrong", and ER (Error Recovery) — the ability to actually fix it. This maps directly to the monitoring-control model of Nelson & Narens (1990) in cognitive psychology.
Three Findings Across 9 SOTA Models
We evaluated GPT-5.2, Claude Opus 4.6, Gemini 3 Pro, DeepSeek-V3.2, Kimi K2.5, and others across 100 expert-level tasks:
1. ER Dominance. 94.8% of MetaCog gain comes from Error Recovery alone. The bottleneck to AGI is not knowledge or reasoning — it is self-correction.
2. Declarative-Procedural Gap. All 9 models can verbalize uncertainty (MA = 0.694) but cannot act on it (ER = 0.302). They sound humble but fail to self-correct — the most dangerous AI safety profile.
3. Difficulty Effect. Harder tasks benefit dramatically more from metacognition (Pearson r = -0.777, p < 0.001).
Paper: FINAL Bench: Measuring Functional Metacognitive Reasoning in LLMs
FINAL Bench is the first tool to tell apart what AI truly knows from what it merely pretends to know.
We release FINAL Bench, the first benchmark for measuring functional metacognition in LLMs — the ability to detect and correct one's own reasoning errors. Every existing benchmark measures final-answer accuracy. None measures whether AI knows it is wrong.
Dataset: [FINAL-Bench/Metacognitive]( FINAL-Bench/Metacognitive) | 100 Tasks | 15 Domains | 8 TICOS Types | Apache 2.0
Leaderboard: FINAL-Bench/Leaderboard
Article: https://huggingface.co/blog/FINAL-Bench/metacognitive
Core Innovation
Our 5-axis rubric separates what no prior benchmark could: MA (Metacognitive Accuracy) — the ability to say "I might be wrong", and ER (Error Recovery) — the ability to actually fix it. This maps directly to the monitoring-control model of Nelson & Narens (1990) in cognitive psychology.
Three Findings Across 9 SOTA Models
We evaluated GPT-5.2, Claude Opus 4.6, Gemini 3 Pro, DeepSeek-V3.2, Kimi K2.5, and others across 100 expert-level tasks:
1. ER Dominance. 94.8% of MetaCog gain comes from Error Recovery alone. The bottleneck to AGI is not knowledge or reasoning — it is self-correction.
2. Declarative-Procedural Gap. All 9 models can verbalize uncertainty (MA = 0.694) but cannot act on it (ER = 0.302). They sound humble but fail to self-correct — the most dangerous AI safety profile.
3. Difficulty Effect. Harder tasks benefit dramatically more from metacognition (Pearson r = -0.777, p < 0.001).
from datasets import load_dataset
dataset = load_dataset("FINAL-Bench/Metacognitive", split="train")Paper: FINAL Bench: Measuring Functional Metacognitive Reasoning in LLMs
FINAL Bench is the first tool to tell apart what AI truly knows from what it merely pretends to know.
- 6 replies
upvoted an article 6 months ago
Article
GGML and llama.cpp join HF to ensure the long-term progress of Local AI
- +4
ggerganov, ngxson, allozaur, lysandre, victor, julien-c
• • 510 liked a
model 6 months ago
