Models & Labs
Gemini 4 Carbon Launches for General Access
Sam Witteveenhigh confidence
Why it matters
- →Provides a dedicated model for software engineering tasks within the Gemini lineup.
- →Demonstrates improved capability in generating complete applications from single prompts.
- →Expands access to Google's latest coding-optimized AI technology for general users.
Google has made Gemini 4 Carbon available for general access, marking its entry into the consumer and developer market. The model is designed specifically for software engineering tasks, aiming to outperform previous iterations like Argon in code generation and application building. Google highlights four key capabilities focused on coding efficiency and includes demos showing the creation of functional apps from simple prompts. This release expands the Gemini family's reach, providing developers with a new tool optimized for practical development workflows.
Read originalThe story around this
Earlier coverage that leads up to this article, and what followed. Lines connect each piece to the closest one after it, converging here.
How we got here
- 1Google's August AI Innovations: Gemini and Beyond
Google AI Blog · September 1, 2026 · Related
- 2Google Launches Gemini 3.8 Flash Model
The Verge AI · September 2, 2026 · Same story
- 3Google Releases Gemini 3.8 Flash Model
Matthew Berman · September 3, 2026 · Related
- 4Google releases Gemini 4 Argon with restricted access
The Verge AI · September 30, 2026 · Same story
- 5Google Gemini 4 Argon and OpenAI DevDay announcements
Wes Roth · October 1, 2026 · Same story
- 6Google Announces Gemini 4 Argon with 1M Token Output
Sam Witteveen · October 1, 2026 · Same story
- 7Gemini 4 Argon arrives in GitHub Copilot
GitHub Changelog · October 13, 2026 · Same story
- 8Google Gemini 4 Argon Benchmarks and Coding Performance
The AI Daily Brief · October 14, 2026 · Same story
What happened next
- 9Google Releases Gemini 4 Carbon Model
The AI Daily Brief · November 12, 2026 · Same story
- 10Google deploys Gemini 4 Hydrogen for faster Search
TechCrunch AI · November 12, 2026 · Related
Follow this story
Open the full story →Google launches Gemini 4 Carbon for cybersecurity
11 developments
- Oct 28 · The Verge AI
Google launches Gemini 4 Carbon for cybersecurity - Oct 29 · Matthew Berman
Google Gemini 4 Carbon wins benchmarks↳ Gemini 4 Carbon wins recent benchmark leaderboards. - Oct 29 · The Rundown AI
Google Gemini 4 Carbon tops benchmarks - Oct 30 · The AI Daily Brief
Google's Gemini 4 Carbon Tops New AI Leaderboard - Oct 30 · Lev Selector
Google Gemini 4 Carbon Leads LM Arena Leaderboard↳ Gemini 4 Carbon tops the LMSYS Chatbot Arena leaderboard. - Oct 30 · Matt Wolfe
Google Releases Gemini 4 Carbon for Cost-Efficient Inference↳ Gemini 4 Carbon is optimized for cost-efficient inference tasks. - Nov 2 · AI Explained
Google Launches Gemini 4 Carbon for Efficiency - Nov 10 · The Verge AI
Google launches Gemini 4 Carbon with agent swarm↳ Gemini 4 Carbon includes an agent swarm feature for orchestrating multiple agents from a single prompt. - Nov 10 · Sam Witteveen
Gemini 4 Carbon Launches for General Access (This article) - Nov 10 · GitHub Changelog
Gemini 4 Carbon arrives in GitHub Copilot↳ Gemini 4 Carbon is now available in GitHub Copilot for developers. - Nov 10 · AI Explained
Google Launches Gemini 4 Carbon for High-Throughput Tasks
More in Models & Labs
Models & Labsother
llama.cpp b12340 adds CUDA 13 and ROCm 10.0
This release quietly closes the hardware gap for local inference by adding default support for CUDA 13 and ROCm 10.0 across Linux and Windows. Users with newer NVIDIA GPUs or AMD accelerators no longer need to compile from source to get performance, while Apple Silicon builds now ship KleidiAI disabled by default, shifting the focus back to standard ARM optimizations. The inclusion of Snapdragon support for Linux and Android marks a significant step toward on-device AI beyond the x86/ARM desktop ecosystem. It is less about new features and more about ensuring llama.cpp remains the universal runtime for every available accelerator.
llama.cpp Releases
Models & Labscoding
llama.cpp b12342 adds few-row MMQ for speculative decoding
Speculative decoding just got a significant speed boost on NVIDIA GPUs. This release introduces a specialized kernel for 'few-row' matrix multiplications, targeting the small batch sizes typical of verification passes in speculative decoding workflows. By optimizing these specific operations for quantized formats like Q8_0 and IQ4 variants, llama.cpp reduces latency where it matters most during token generation. The change is CUDA-only for now but leverages hardware capabilities from Ampere onwards, making local inference faster without requiring new model architectures or complex configuration changes.
llama.cpp Releases
© TechCrunch AIModels & Labsmodels
Google deploys Gemini 4 Hydrogen for faster Search
Google is quietly shifting its search infrastructure by replacing older fast models with Gemini 4 Hydrogen, a tier designed specifically for speed and cost-efficiency rather than raw reasoning power. This model completes tasks 2.5 times faster than its predecessor while beating the more powerful Argon on specific intelligence benchmarks, proving that specialized efficiency can outperform brute force in high-volume workflows. By integrating this into Search's fan-out technique, Google now breaks down complex queries into subqueries with greater precision and speed, delivering richer visual responses without the latency of heavier models. This marks a strategic pivot toward agentic execution layers within consumer search, making AI Mode feel snappier for everyday multi-part questions.
TechCrunch AI