|
Running small LLMs locally — what hardware are you using in 2026?
|
|
1
|
55
|
October 14, 2026
|
|
Unable to select Python version 3.13 on Spaces
|
|
1
|
29
|
October 14, 2026
|
|
Wan2gp best setting for fasteste generation
|
|
1
|
38
|
October 14, 2026
|
|
openTPU: running Hugging Face language models on an open-source FPGA accelerator
|
|
0
|
30
|
October 13, 2026
|
|
How to keep what the model makes up out of the agent
|
|
7
|
100
|
October 14, 2026
|
|
Picking STS3215 servos for SO-101: am I the only one confused?
|
|
1
|
56
|
October 13, 2026
|
|
Local AI - How do you choose your models
|
|
3
|
76
|
October 14, 2026
|
|
Useful parameter for Model/Dataset/Space search
|
|
3
|
146
|
October 10, 2026
|
|
Do you Think Removing the $0.10 Monthly Free Inference Credits Is the Right Decision?
|
|
2
|
190
|
October 9, 2026
|
|
Technical scaling & spatial fit for SKU object replacement
|
|
1
|
67
|
October 13, 2026
|
|
How do you reproduce LLM agent failures? I built an open-source Python tool for offline replay and pytest testing
|
|
4
|
96
|
October 14, 2026
|
|
Fitting Gemma 4 E2B and an SDXL model on one iPhone: notes from building Anvil
|
|
0
|
28
|
October 13, 2026
|
|
We gave AI a trust we never gave people - The third axis, after capability and permission
|
|
10
|
482
|
October 8, 2026
|
|
Shoe string ablation or pruning
|
|
12
|
238
|
October 13, 2026
|
|
Scaling Beatrix V3: a 376M byte-level model and the training
|
|
9
|
204
|
October 11, 2026
|
|
What does a AI judge score of 0.8 actually mean? Introducing Typed Evals
|
|
2
|
106
|
October 14, 2026
|
|
What do you think about AI?
|
|
21
|
983
|
October 9, 2026
|
|
Could RAG retrieve LLM-native latent memory instead of text?
|
|
7
|
153
|
October 14, 2026
|
|
How to improve my tokens per second?
|
|
1
|
126
|
October 10, 2026
|
|
Sanity check on my hybrid RAG settings
|
|
3
|
85
|
October 13, 2026
|
|
Did HuggingFace remove the credits given to Free Users?
|
|
5
|
346
|
October 10, 2026
|
|
Anyone else having trouble getting a ZeroGPU allocation/reservation?
|
|
37
|
1875
|
October 13, 2026
|
|
Coordinated spam wave flooding the "New" topics feed
|
|
2
|
143
|
October 12, 2026
|
|
Janas-LLM 37 tok/s vs llama.cpp 2.8 tok/s on Qwen3-Next-80B, on an Intel Core Ultra 9 185H with 32 GB RAM
|
|
0
|
101
|
October 11, 2026
|
|
Estimating local decode speed of MoE models from active parameters: method and calibration
|
|
1
|
42
|
October 14, 2026
|
|
Keeping the Attention Sink Exact in Activation-Whitened SVD Compression
|
|
2
|
112
|
October 11, 2026
|
|
Rdq: does a robot dataset's licence match its source? (checked 3,000 LeRobot datasets)
|
|
0
|
30
|
October 12, 2026
|
|
An open research lab for measuring — not prompting — music production in a DAW
|
|
14
|
415
|
October 10, 2026
|
|
Negative result: tail-latency gating for KV offload doesn't beat LRU+TTL (runnable simulator)
|
|
0
|
29
|
October 12, 2026
|
|
The Confidence Trap: Why AI Systems Are Deliberately Trained to Sound Certain When They're Wrong
|
|
9
|
253
|
October 10, 2026
|