Search: huggingface.co | Heykuki News

HK

Heykuki News

Top New Best Ask Show Jobs

Top New Best Ask Show Jobs

661.

Open LLM Leaderboard

3 years ago

6 points

662.

RWKV – An RNN with the Advantages of a Transformer

3 years ago

6 points

663.

Open LLM Leaderboard

3 years ago

6 points

664.

Assisted Generation: a new direction toward low-latency text generation

3 years ago

6 points

665.

Databricks Publishes a Version of Dolly LLM to Hugging Face

3 years ago

6 points

666.

Hugging Face introduces Pull Requests and Discussions

4 years ago

6 points

667.

Show HN: KaniTTS – Open-source high-fidelity TTS with just 450M params

9 months ago

5 points

668.

First DeepSeek V4 Flash-Base-Int4 Quant

2 months ago

5 points

669.

APIEval-20: A Benchmark for Black-Box API Test Suite Generation

3 months ago

5 points

670.

a year ago

5 points

671.

An open source common knowledge and context based Hallucination Detection Model

a year ago

5 points

672.

HuggingFace Is Down

status.huggingface.co

2 years ago

5 points

673.

TinyLlama a 1.1B Llama model trained on 3T tokens reaches 1.0 release

2 years ago

5 points

674.

Gemma 3 QAT (Quantized Aware Training) 3x less memory

a year ago

5 points

675.

9 days ago

5 points

676.

I made a CPU only spiking neuron network lib that comes pretty close to PyTorch

a month ago

5 points

677.

3 months ago

5 points

678.

Do Bubbles Form When AIs Simulate Capitalism?

4 months ago

5 points

679.

Craft – image models can think like LLMs

5 months ago

5 points

680.

Sopro v1.5: A 135M TTS model trained for ~$100, runs 20× real-time on CPU

5 months ago

5 points

681.

HuggingFace Skills: Fine-tune any LLM with one sentence for $0.30

7 months ago

5 points

682.

Show HN: Dante-Qwen-4B – Curing LLM "Neurosis" with a Divine Comedy Curriculum

7 months ago

5 points

683.

The Smol Training Playbook: The Secrets to Building World-Class LLMs

8 months ago

5 points

684.

Open source 3B param model for documents better than Gemini 2.5

8 months ago

5 points

685.

Gaia2 and Are: Empowering the Community to Evaluate Agents

9 months ago

5 points

686.

SWE-rebench: Over 21,000 Open Tasks for SWE LLMs

a year ago

5 points

687.

Mixture of Tunable Experts-DeepSeek R1 Behavior Modification at Inference Time

a year ago

5 points

688.

DeepSeek-V3-0324

a year ago

5 points

689.

Perplexity post-trained deepseek-r1 to remove Chinese Communist Party censorship

a year ago

5 points

690.

Moonshine – open-source, real-time speech-to-text in the browser

2 years ago

5 points