30 articles · page 1 of 1
GLM-5.3-Flash Intelligence, Performance and Price Analysis
2026-08-26
New Mac Studio with M5 Max and M5 Ultra
2026-08-25
How we made a text-to-speech model respond in sub-50 ms
2026-08-21
GLM-5.3 Artificial Analysis Benchmarks
2026-08-18
Introducing Toast 1
2026-08-14
Accelerating GPT-5.6 Sol Ultrafast
2026-08-13
DeepSeek V4 Pro 0813
2026-08-12
Automatic1111 for Apple metal, 40% speed up sd1.5
2026-08-12
DeepSeek V4 Flash 0731
2026-08-07
DeepSeek V4 Flash 0731 Intelligence, Performance and Price Analysis
2026-07-31
Performance per dollar is getting faster and cheaper
2026-07-03
Popping the GPU Bubble
2026-06-30
GLM 5.2 Performance Benchmarks
2026-06-17
DeepSeek V4 Pro beats GPT-5.5 Pro on precision
2026-06-08
Nvidia RTX Spark
2026-06-01
Arena AI Model ELO History
2026-05-14
Lambda Calculus Benchmark for AI
2026-04-25
Claude Opus 4.7 costs 20–30% more per session
2026-04-17
Claude Opus 4.6 accuracy on BridgeBench hallucination test drops from 83% to 68%
2026-04-12
A leak reveals that Anthropic is testing a more capable AI model "Claude Mythos"
2026-03-27
Quantization from the Ground Up
2026-03-25
MacBook M5 Pro and Qwen3.5 = Local AI Security System
2026-03-20
Are LLMs not getting better?
2026-03-12
LLMs work best when the user defines their acceptance criteria first
2026-03-07
Scientists made AI agents ruder — and they performed better at complex reasoning tasks
2026-03-02
Unsloth Dynamic 2.0 GGUFs
2026-02-28
Fast KV Compaction via Attention Matching
2026-02-20
Consistency diffusion language models: Up to 14x faster, no quality loss
2026-02-20
Improving 15 LLMs at Coding in One Afternoon. Only the Harness Changed
2026-02-12
Anthropic's original take home assignment open sourced
2026-01-21