tokens

Meta unleashes Llama API running 18x faster than OpenAI: Cerebras partnership delivers 2,600 tokens per second

When AI reasoning goes wrong: Microsoft Research shows more tokens can mean more problems

OpenAI’s new GPT-4.1 models can process a million tokens and solve coding problems better than ever

DeepSeek-V3 Now Runs At 20 Tokens Per Second On Mac Studio

DeepSeek-V3 now runs at 20 tokens per second on Mac Studio, and that’s a nightmare for OpenAI

Locks, leases, fencing tokens, FizzBee

Qwen2.5-1M: Deploy your own Qwen with context length up to 1M tokens

Malicious PyPi package steals Discord auth tokens from devs

Meta’s new BLT architecture replaces tokens to make LLMs more efficient and versatile

Byte Latent Transformer: Patches Scale Better Than Tokens

Hyrumtoken: A Go package to encrypt pagination tokens

Llama 3.1 405B now runs at 969 tokens/s on Cerebras Inference

Cerebras Inference now 3x faster: Llama3.1-70B breaks 2,100 tokens/s

Trump Crypto Project Website Crashes as Its Tokens Go on Sale

Llama 405B 506 tokens/second on an H200

The Role of Anchor Tokens in Self-Attention Networks

Elixir Games allows users to stake tokens directly within its launcher

Donald Trump is hawking tokens for a crypto project he still hasn’t explained

500,000 tokens: How Anthropic’s Claude Enterprise is pushing AI boundaries

Cerebras launches inference for Llama 3.1; benchmarked at 1846 tokens/s on 8B