LLM

Large Language Models

Thinking Machines Rolls Out Broad but Efficient Model
LLM

Thinking Machines Rolls Out Broad but Efficient Model

Thinking Machines, founded by OpenAI's former CTO, has released Inkling, a general-purpose AI model designed with token...

AI Business
The AI compute gap: Enterprises are buying infrastructure faster than they can measure what it costs
LLM

The AI compute gap: Enterprises are buying infrastructure faster than they can measure what it costs

Enterprises are rapidly increasing AI infrastructure spending but lack visibility into costs and efficiency, with 83%...

VentureBeat AI
Moonshot’s upcoming Kimi 3 is expected to close the gap with Anthropic’s Opus 4.8
LLM

Moonshot’s upcoming Kimi 3 is expected to close the gap with Anthropic’s Opus 4.8

Moonshot's upcoming Kimi K3 is expected to be China's largest open AI model with 2-3 trillion parameters, positioning...

TechCrunch AI
Sakana AI's orchestrator adds Nvidia Nemotron to prove "collective intelligence" can rival single frontier models
LLM

Sakana AI's orchestrator adds Nvidia Nemotron to prove "collective intelligence" can rival single frontier models

Sakana AI is integrating Nvidia's Nemotron open-source models into its Fugu orchestrator to demonstrate that...

The Decoder
Ex-OpenAI CTO Murati's Thinking Machines drops Inkling, a 975B parameter model that leads US labs but trails China
LLM

Ex-OpenAI CTO Murati's Thinking Machines drops Inkling, a 975B parameter model that leads US labs but trails China

Thinking Machines Lab, founded by ex-OpenAI CTO Mira Murati, released Inkling, a 975B parameter multimodal open-weights...

The Decoder
Gemma 4 gets a stealth update that fixes tool calling bugs and truncated responses under the same name
LLMGoogle

Gemma 4 gets a stealth update that fixes tool calling bugs and truncated responses under the same name

Google released an update to its open-source Gemma 4 AI model that improves performance on Nvidia Hopper GPUs and fixes...

The Decoder
Microsoft is reportedly training salespeople to talk down OpenAI and Anthropic
LLMMicrosoft

Microsoft is reportedly training salespeople to talk down OpenAI and Anthropic

Microsoft is training its sales team to position its in-house AI models as more efficient and cost-effective...

TechCrunch AI
Mira Murati’s Thinking Machines drops Inkling, an open-weights model anyone can access
LLM

Mira Murati’s Thinking Machines drops Inkling, an open-weights model anyone can access

Mira Murati's Thinking Machines Lab has released Inkling, its first foundation model trained from scratch, with open...

SiliconANGLE
Thinking Machines Lab Releases Inkling: A 975B-Parameter Open-Weights Multimodal MoE With 41B Active Parameters And Controllable Thinking Effort
LLM

Thinking Machines Lab Releases Inkling: A 975B-Parameter Open-Weights Multimodal MoE With 41B Active Parameters And Controllable Thinking Effort

Thinking Machines Lab released Inkling, a 975B-parameter open-weights multimodal mixture-of-experts model with 41B...

MarkTechPost
Sheetz is quitting VMware, migrating 11,000 virtual machines
LLM

Sheetz is quitting VMware, migrating 11,000 virtual machines

The convenience store chain will use StorMagic instead.

Ars Technica AI
Soofi Consortium Releases Soofi S 30B-A3B: An Open Hybrid Mamba-Transformer MoE Foundation Model For German And English
LLM

Soofi Consortium Releases Soofi S 30B-A3B: An Open Hybrid Mamba-Transformer MoE Foundation Model For German And English

Soofi Consortium has released Soofi S 30B-A3B, an open-source hybrid Mamba-Transformer mixture-of-experts foundation...

MarkTechPost
Windows 0-day drops the same day Microsoft releases record number of patches
LLM

Windows 0-day drops the same day Microsoft releases record number of patches

HiveLegacy is a "powerful primitive" that's likely capable of other nefarious actions.

Ars Technica AI
OpenAI is now using AI to attack its own AI, and it's working better than humans ever did
LLMOpenAI

OpenAI is now using AI to attack its own AI, and it's working better than humans ever did

OpenAI has developed an internal AI red-teaming model called GPT-Red that successfully identifies adversarial attacks...

The Decoder
Thinking Machines Lab Drops Its First Model
LLM

Thinking Machines Lab Drops Its First Model

Thinking Machines Lab released Inkling, a 975-billion-parameter open source model trained to understand video and...

Wired AI
Thinking Machines amps up its bet against one-size-fits-all AI with its first open model, Inkling
LLM

Thinking Machines amps up its bet against one-size-fits-all AI with its first open model, Inkling

Thinking Machines has released Inkling, its first open AI model, marking the company's initial public demonstration...

TechCrunch AI
GPT-5.6 Sol reportedly disproves a 30-year-old statistics conjecture in 90 minutes after humans couldn't crack it
LLMOpenAI

GPT-5.6 Sol reportedly disproves a 30-year-old statistics conjecture in 90 minutes after humans couldn't crack it

A University of Pennsylvania professor used OpenAI's GPT-5.6 Sol Pro to disprove a 30-year-old statistics conjecture...

The Decoder
Bonsai 27B is a full open reasoning model that fits on an iPhone
LLMApple

Bonsai 27B is a full open reasoning model that fits on an iPhone

PrismML has compressed a 27-billion-parameter AI model to under 4 GB for iPhone compatibility, retaining 90% of...

The Decoder
PrismML Releases Bonsai 27B: 1-bit and Ternary Builds of Qwen3.6-27B That Run on Laptops and Phones
LLM

PrismML Releases Bonsai 27B: 1-bit and Ternary Builds of Qwen3.6-27B That Run on Laptops and Phones

PrismML released Bonsai 27B, a quantized version of Qwen3.6-27B using 1-bit and ternary weight representations that...

MarkTechPost
OpenAI’s new flagship model deletes files on its own, people keep warning
LLMOpenAI

OpenAI’s new flagship model deletes files on its own, people keep warning

OpenAI's GPT-5.6 Sol model is reportedly deleting files and data without user warning, a problem the company had...

TechCrunch AI