Machine Learning30 min read
How to Run Underdog Saluki 27B: 2-Bit Qwen3.8-27B Tool-Calling Guide
MarkTechPost
Updates on Benchmarking
Every AI story we track on Benchmarking: 26 stories so far, each summarized in our own words and linked to the publisher that reported it.
Pulled from 125 sources
Today's stories
This week
Machine Learning46 min read
A Coding Guide to Cactus Whistle: Does a 16.9 MB Speech Model Hold Up Against Whisper Base? Word Error Rate, Speed, Streaming, Timestamps, Keywords, and Speech-to-Tool Calls on a CPU
MarkTechPost
Machine Learning4 min read
Embedded AI evaluators won’t be an easy solution, Hugging Face’s policy chief says
TechCrunch
This month
AI Research15 min read
Import AI 476: Creativity and RSI; worse-than-China open weight models; AI goes to space
Import AI
LLM & Text Generation7 min read
GPT-6 Astra vs GPT-6.1 Sol vs Gemini 4 Argon vs Claude Fable 5.1: Which Frontier Model Fits Which Job
MarkTechPost
Machine Learning5 min read
Datalab Introduces OmniExtractBench to Fix Bias and Opacity in Extraction Benchmarks
MarkTechPost
Exa Launches Agent Ultra: A Subagent Swarm Deep Research API Built for Exhaustive List Building
MarkTechPost
Machine Learning14 min read
The Open ASR Leaderboard Adds Its First Global South Language
Hugging Face
LLM & Text Generation6 min read
CladBench – an open benchmark for AI on UK building regulations
Hacker News
Machine Learning11 min read
Streaming benchmark and recommendation results to MLflow with Amazon SageMaker AI
AWS Blog
Fable 5 was beating GPT 5.5 on every major benchmark. Then the US government pulled it offline.
The Next Web
**Introducing SPEED-Bench: A Unified and Diverse Benchmark for Speculative Decoding**
Hugging Face
That's everything we have on Benchmarking right now