218 summarised stories about Model Releases, each linking back to the original source. Browse all topics →
Zvi (Don't Worry About the Vase) 1 day ago 18 ● 4 sources
Scion, Anthropic’s internal GitHub replacement, is said to be driving an exponential rise in the amount of code Anthropic produces. Anthropic Q3 revenue is reported as $16.8b. The article also covers newly released model reviews and releases including Google Gemini 4 Argon and Meta Muse Wonder, but says full capability and safety coverage will depend on how well those models perform.
Hugging Face 1 day ago 3
Granite Switch 4.1 combines decision and speech adapters into one hosted model endpoint so it can handle chat, decision requests, and speech-to-text tasks. Granite-Speech-5.0-Turbo-CTC is claimed to be the fastest ASR model on the OpenASR leaderboard. The release reuses the base model’s KV cache across adapters and aims to deliver more capabilities without additional GPU memory beyond what the base model alone would use.
The New Stack 2 days ago 37
OpenAI will retire GPT-4o, GPT-4o mini, GPT-4.1, and GPT-4.1 mini from its API on January 14, 2027, including dated snapshots and fine-tunes based on them. The deprecation notice delivered Wednesday gives developers about 3 months to migrate production traffic. Developers must move to GPT-6 Luna and adjust API usage and settings because tool-calling and reasoning behavior differs, including limits on temperature/top_p/logprobs and the need to use the Responses API for combined reasoning plus tools.
Mistral 2 days ago 22
Mistral released Timestral, its first family of time-series foundation models, and reports top results on major public forecasting benchmarks with probabilistic, multivariate forecasts. The largest model has 209M parameters and ranks first on GIFT-Eval, TIME, and fev-bench. The release introduces four model sizes (5M to 209M) designed for accuracy/efficiency tradeoffs, CPU inference, and zero-shot forecasting, with optional fine-tuning through Mistral Forge and code made available on GitHub.
Product Hunt 2 days ago 14 ● 5 sources
NVIDIA launched Nemotron 3.5 Super VL, an open multimodal model for specialized agents. It is a 120B mixture-of-experts model with 12B active parameters. NVIDIA says it can deliver up to 5x higher agentic throughput on a single GPU than open models of similar size, with checkpoints and training materials provided on Hugging Face.
SiliconANGLE 2 days ago 2 ● 12 sources
Anthropic disclosed ahead of its IPO that it generated $16.8 billion in Q3 revenue and reached adjusted profitability for the first time. The company reported a 45% sequential revenue increase in the quarter. The disclosure adds new financial specifics investors can use as it moves toward an expected November IPO, influencing expectations around its fundraising needs and role as a market benchmark for AI companies.
The daily briefing
TLDRocket reads all relevant sources, removes duplicate coverage, and summarises the day in two minutes. Follow companies and topics for alerts, or get the briefing in Slack. Free, no spam, unsubscribe anytime.
TLDRocket
17 October 2026
The big thread today is the messy middle between agent ambitions and the boring parts that make them safe and auditable. ServiceNow is positioning its AI Control Tower as the enterprise hub to discover, govern, and—crucially—control AI agents across models and clouds, while a field study of 15 large-enterprise customers (conducted September 15–21, 2026) presses on a key question: how strong is “technical runtime enforcement” and lifecycle evidence compared with other tools? ServiceNow’s answer is mostly commercial muscle—aiming for $1.5B in full-year AI ACV—but its guidance effectively says buyers should treat Control Tower as strongest for governance workflows and operational action, then verify where model validation evidence and real-time blocking are actually handled.
Read the full briefing →