Models & Labs
Claude Opus 5.5 and GPT-6 Sol Released
Lev Selectorhigh confidence
Why it matters
- →Latest frontier models from top two labs.
- →Improvements in reasoning and coding benchmarks.
- →Increases competition in enterprise AI services.
Anthropic has released Claude Opus 5.5, its latest flagship model focusing on enhanced reasoning and coding abilities. Concurrently, OpenAI introduced the GPT-6 Sol and Luna models, expanding its family of specialized large language models. These releases represent incremental but significant improvements in multimodal understanding and task execution. The updates reinforce the rapid iteration pace among leading AI labs.
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
- 1Claude Opus 4.8 Released by Anthropic
Matt Wolfe · May 29, 2026 · Related
- 2Anthropic Releases Claude Opus 4.8 AI Model
AI Explained · May 29, 2026 · Related
- 3Anthropic Launches Claude Opus 5 Model
Anthropic · July 24, 2026 · Same story
- 4Anthropic Launches Claude Opus 5 Model
The Rundown AI · July 27, 2026 · Same story
- 5Anthropic's Opus 5 Model Surpasses Expectations
The Rundown AI · July 27, 2026 · Related
- 6Claude Opus 5 Challenges Fable and GPT
The AI Daily Brief · July 28, 2026 · Related
- 7Anthropic Releases Claude Opus 5
Matt Wolfe · July 31, 2026 · Same story
- 8OpenAI releases GPT-6 Sol and Luna with lower costs
TechCrunch AI · September 22, 2026 · Related
What happened next
- 9Anthropic Releases Claude Haiku 5.5 for Cost Efficiency
The AI Daily Brief · October 9, 2026 · Related
- 10Anthropic Releases Claude Opus 5.5 System Card
AI Explained · October 12, 2026 · Related
- 11Anthropic ships Claude 3.5 updates and new apps
Duncan Rogoff · October 21, 2026 · Related
- 12Anthropic launches Claude Duet AI agent
The Verge AI · October 21, 2026 · Related
- 13Claude Fable 5.2 arrives in GitHub Copilot
GitHub Changelog · November 9, 2026 · Related
Follow this story
Open the full story →Anthropic releases Claude Opus 5.5 with tighter security
10 developments
- Sep 22 · The Verge AI
Anthropic releases Claude Opus 5.5 with tighter security - Sep 22 · TechCrunch AI
Anthropic releases Opus 5.5 with lower prices↳ Opus 5.5 lowers output costs to $20 per million tokens while maintaining performance that beats the larger Fable model in coding and knowled - Sep 22 · GitHub Changelog
Claude Opus 5.5 arrives in GitHub Copilot↳ Claude Opus 5.5 is now available in GitHub Copilot for enterprise developers. - Sep 23 · Matthew Berman
Anthropic releases Opus 5.5 - Sep 23 · Duncan Rogoff
Anthropic releases official Claude Code playbook for Opus 5.5↳ Anthropic released an official guide optimizing Claude Code with Opus 5.5 to reduce latency and cost through specific prompting habits. - Sep 24 · AI Explained
Anthropic Releases Claude Opus 5.5 Model - Sep 25 · Lev Selector
Claude Opus 5.5 and GPT-6 Sol Released (This article)↳ OpenAI launches GPT-6 Sol and Luna variants alongside Anthropic's release. - Sep 26 · Matt Wolfe
Anthropic Releases Claude Opus 5.5 - Sep 30 · Matt Wolfe
Anthropic Releases Opus 5.5 Coding Model - Oct 2 · Lev Selector
Anthropic Launches Claude Sonnet 5.5 and Opus 5.5
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