- H-FARM AI's Newsletter
- Posts
- Claude Opus 5 ships with perfect Math Olympiad score
Claude Opus 5 ships with perfect Math Olympiad score
PLUS: NVIDIA SANA-Video 2.0 runs 720p on one GPU & ChatGPT Work agent logs into authenticated apps. NVIDIA leads 50-company push for open-weight AI, YouTube Studio rolls out AI thumbnail generator.

1️⃣ Anthropic launches Claude Opus 5 with a perfect 42/42 on the Math Olympiad benchmark and $5/$25 per million token pricing — half the cost of comparable frontier models 2️⃣ NVIDIA Labs releases SANA-Video 2.0, generating up to 720p long-form video on a single GPU in 5B and 14B parameter variants 3️⃣ OpenAI expands the ChatGPT Work agent to interact with authenticated websites, opening access to CRMs, HR tools, and gated enterprise platforms |
|
MAIN AI UPDATES / 27th July 2026
🤖 Claude Opus 5 ships with perfect Math Olympiad score 🤖
Anthropic's new frontier model tops every major benchmark at half rivals' pricing.
Anthropic has launched Claude Opus 5 across its consumer apps, Claude Code, and the API, delivering new state-of-the-art results in coding, reasoning, agentic search, and computer use. The model scored a perfect 42/42 on the 2026 International Math Olympiad benchmark, hit 30.2% on ARC-AGI-3 — triple the next best model — and claimed the top spot on Artificial Analysis' Intelligence Index leaderboard. Priced at $5/$25 per million input/output tokens, Opus 5 approaches Fable 5-level intelligence at roughly half the cost, a competitive pressure move that forces rivals to respond on both capability and pricing. This positions Anthropic as the clear cost-efficiency leader at the frontier.
🎬 NVIDIA SANA-Video 2.0 runs 720p on one GPU 🎬
New architecture slashes hardware requirements for long-form AI video generation.
NVIDIA Labs has released SANA-Video 2.0, a video generation architecture that combines linear attention with periodic softmax layers to produce videos up to 720p resolution on a single GPU. Available in 5B and 14B parameter variants, the models maintain competitive visual quality while cutting latency for long-form, high-resolution output. By eliminating multi-GPU requirements, SANA-Video 2.0 could reshape the pricing and access economics of AI video generation, enabling broader adoption across creative and enterprise workflows. This matters because it lowers the barrier to entry for studios and solo creators alike, making high-quality AI video practical without expensive infrastructure.
🔐 ChatGPT Work agent logs into authenticated apps 🔐
OpenAI's agent can now access gated enterprise tools, expanding rollout into real workflows.
OpenAI has expanded the ChatGPT Work agent to support websites requiring user authentication, enabling the agent to complete tasks across gated apps and services on the user's behalf. The update transforms the Work agent from a general web assistant into a capable digital worker that can interact with CRMs, project management tools, HR systems, and other login-protected platforms. This is a prerequisite for meaningful enterprise automation, and it positions OpenAI more aggressively in the enterprise agent space where authenticated access is table stakes for real adoption.
INTERESTING TO KNOW
📜 NVIDIA leads 50-company push for open-weight AI 📜
NVIDIA, Microsoft, Meta, and 47 other tech companies, VCs, and nonprofits signed an open letter urging Washington not to restrict open-weight AI models, with OpenAI, Google, AMD, and Cisco subsequently joining — though Anthropic notably did not. The coalition argues open weights preserve competition, broaden access, and prevent vendor lock-in. The regulation debate now splits the frontier AI industry along a clear philosophical divide on openness versus safety-driven restrictions, and it could shape upcoming US AI policy.
🎨 YouTube Studio rolls out AI thumbnail generator 🎨
YouTube Studio is rolling out AI-powered custom thumbnail generation through its integrated chatbot, letting creators generate and iterate on thumbnails directly within the platform. Thumbnails are among the most critical factors for click-through rates, making this a high-impact integration for YouTube's millions of active creators. The tool leverages Google's image generation capabilities and intensifies competitive pressure on standalone AI design tools targeting the creator economy.

📩 Have questions or feedback? Just reply to this email , we’d love to hear from you!
🔗 Stay connected:
