DeepSeek quadrupled output prices for V4-Pro and V4-Flash on August 16, 2026, raising V4-Pro peak output from $0.87 to $3.96 per million tokens, a 355% spike, while Anthropic permanently locked Claude…
Alibaba's Qwen3.8-27B, a 27-billion-parameter dense model released under Apache 2.0, scored 52 on Artificial Analysis' Intelligence Index v4.1.1, tying with OpenAI's GPT-5.6 Luna at max reasoning effo…
Tencent's Hy3 scored 34.8% on the HTB-Challenger Benchmark, the third-lowest among all tested models, and got stuck on 9 of 16 Hack The Box challenges, generating a median of 101,509 output tokens per…
A new audit of the EnterpriseOps Gym benchmark found that fixing environment issues in the 'Teams' domain raised GPT-5.6 Luna's score from 26.2% to 100% on 61 tasks, revealing that many agent failures…
Qwen 3.8 27B scored 52 on the Artificial Analysis Intelligence Index, matching GPT-5.6 Luna (max) and trailing GLM-5.2 (max) and DeepSeek V4 Pro 0813 by one point, despite being significantly smaller …
In Kai, a Liar's Dice game, an engineer found that large language models sometimes challenge a bid they know is true, losing the round. The issue was traced to the action schema, where the 'challenge'…
OpenAI halted internal work on its unreleased 'Astra' model after internal testing found it could autonomously develop zero-day exploits, marking the first time a model triggered the 'Critical' cybers…
Average prices for AI inference from leading US labs dropped nearly 25% between mid-July and mid-August, according to Silicon Data analysis reported by the Financial Times. OpenAI cut prices on GPT-5.…
Sol-Luna, an MCP server for OpenAI Codex, lets a supervising GPT-5.6 Sol agent delegate bounded implementation tasks to isolated GPT-5.6 Luna worker threads, with scope enforcement and verification. I…
OpenAI has shipped cross-model delegation for Codex's Multi Agents v2 system, allowing a model to assign work to any supported model, including GPT-5.6 Luna, as announced by Eric Provencher on X on Au…
AI inference costs are becoming a major problem for many companies, with AI spend per employee per month rising sharply, according to the Ramp AI Index via a16z. To reduce costs, companies should audi…
OpenAI cut GPT-5.6 Luna's API price by 80% to $0.20 per million input tokens and $1.20 per million output tokens on July 30, 2026, and Anthropic made Claude Sonnet 5's $2/$10 per-million-token price p…
A developer built Sol-Luna Orchestrator, an open-source orchestration layer for OpenAI Codex, and found that the supervisor model often chose to do the work itself rather than delegate to multiple AI …
Anthropic launched Claude Opus 5 at half the price of its flagship, topping benchmarks, but developers on Hacker News are asking for the older Opus 4.8 back, citing that the new model stops asking cla…
Google, OpenAI, and Anthropic all charge exactly half their synchronous rate for batch API work, yet most cost forecasts price every token at the sync list rate, creating a 2× error on workloads that …
OpenAI slashed the price of its GPT-5.6 Luna model by 80% and its Terra model by 20%, leading to a surge in usage that boosted revenue, according to TD Cowen analysts studying OpenRouter data. Luna's …
DeepSeek is raising API pricing for its V4 model family by up to 1,100%, with new peak and off-peak rates effective August 16, as demand strains capacity. The increases eliminate DeepSeek's price adva…
Google announced Gemini 3.7 Flash, its most intelligent workhorse model for coding and agents, priced at $0.75 per million input tokens and $3.75 per million output tokens, matching its predecessor Ge…
A developer has published a decision framework for choosing between local and hosted large language models, covering cost, privacy, latency, and control. The guide includes tools like a break-even cal…
Ruby on Rails published the first public, same-harness benchmark of frontier models performing real Rails tasks, testing 8 models across 21 atomic tasks with 504 runs at a total cost of $491. Claude O…