Stories by swyx

LTM-2-Mini is our first model with a 100M token context window

Write your own LLM benchmarks – with Nicholas Carlini

The on‑Device Intelligence Update

H100 Freakonomics: Cracks in the Bubble?

The Vizier Gaussian Process Bandit Algorithm

Beating OpenAI's structured outputs on cost, accuracy and speed

Surfer: Centralize all your personal data from online platforms

Is finetuning GPT4o worth it?

Llamafile v0.8.13 (and Whisperfile)

Justine Tunney on X: "I just launched whisperfile"

DeepSeek-Prover-v1.5: Proof Assistant for RL and Monte-Carlo Tree Search

Shipping 1000s of successful products with no managers and a team of 12

XBOW matches the capabilities of a top human pentester

Genie Scores 30% on SWE-Bench (new SOTA)

Segment Anything 2: Demo-First Model Development

OpenAI: Structured Outputs in the API

Fastcore: An Underrated Python Library

From Live Data to High-Quality Benchmarks: The Arena-Hard Pipeline – Lmsys Org

How Fireworks evaluates quantization precisely and interpretably

OpenAI-backed legaltech startup Harvey raises $100M

1
2
3
4
5
6
7
8
9
10