Claude Code is hiding fingerprints in your terminal output
A security researcher found Claude Code embeds invisible Unicode markers in API requests to classify users by gateway, timezone, and competing AI lab keywords.
NESTFRONTIER / ARCHIVE
All entries
Technical analysis, model releases, research, and deployment notes.
PAGE_13
294 PUBLISHED ENTRIESMeituan's open model matched GPT-5.5. They deliver food for a living.
Meituan's 1.6T open model matched GPT-5.5 on SWE-bench Pro, trained on 50K domestic cards. MIT license, $0.75/M input tokens.
That API call is secretly five models arguing
vLLM Semantic Router turns one API call into a bounded multi-model collaboration. It beats Fugu Ultra on coding benchmarks, and you can run it yourself.
Same resume got 66 to 99. HackerRank's AI hiring is random
HackerRank open-sourced their AI resume scorer. Running the same resume 100 times produced scores from 66 to 99. The code is honest. The scoring is random.
Open-weight AI just beat Claude at finding security bugs
An open-weight model from China just beat Claude at finding security vulnerabilities. The cost per bug found was 17 cents. But the real story is where the model's intelligence came from.
OpenAI built its own chip. Here's why that changes everything
 OpenAI just built its own chip. Not a GPU. Not a TPU. An ASIC designed from scratch for infere...
A 753B open-source model just matched Claude on coding
Zhipu AI's GLM-5.2 is a 753B open-weight MoE model under MIT license that matches Claude Opus on coding benchmarks, with a genuine 1M token context window and $18/month flat-rate pricing.
Your coding agent's harness is the bottleneck it can't see
DeepReinforce's Ornith-1.0 learns to write its own RL scaffolds, and the 35B model beats Qwen 3.5-397B on terminal tasks.
91.9% coding score. The government decides who uses it.
OpenAI shipped GPT-5.6 Sol with a 91.9% coding score. Almost nobody can use it. The US government controls who gets access.
6,000 People Tried to Hack This AI Assistant. Nobody Succeeded.
A developer challenged 2,000 people to hack his AI assistant via email. After 6,000 attempts, nobody got the secrets. Here's what that actually proves about AI security.