Thinking Machines Lab
PulseAugur coverage of Thinking Machines Lab — every cluster mentioning Thinking Machines Lab across labs, papers, and developer communities, ranked by signal.
- founded by Mira Murati 100%
- founded Inkling 95%
- instance of Inkling 95%
- developed Inkling-Small 95%
- developed by Inkling-Small 95%
- developed Inkling 90%
- developed by Inkling 90%
- founded Mira Murati 90%
- founded by Lilian Weng 90%
- employed by Lilian Weng 90%
- developed Thinking machines 90%
- founded Lilian Weng 90%
- 2026-09-17 hiring Yao Xingcheng, former head of pre-training for Tencent's Hunyuan model, has joined Thinking Machines Lab. source
- 2026-06-05 product_launch Mira Murati previewed "interaction models" from Thinking Machines Lab, designed for continuous real-time processing of audio, text, and video. source
- 2026-05-17 product_launch Thinking Machines Lab launched its first AI model. source
- 2026-05-17 product_launch Thinking Machines Lab launched its first AI model focused on human-AI collaboration. source
- 2026-05-15 product_launch Thinking Machines Lab previewed new "interaction models" designed for continuous human communication.
- 2026-05-12 research_milestone Thinking Machines Lab unveiled its vision for full-duplex conversational AI, which closely resembles prior work from 面壁智能. source
- 2026-05-12 product_launch Thinking Machines Lab introduced interaction models, a new architecture for real-time human-AI collaboration. source
- 2026-05-12 product_launch Thinking Machines Lab introduced interaction models, a new type of AI for real-time collaboration. source
- 2026-05-12 product_launch Thinking Machines Lab announced its new interaction models, featuring full-duplex AI capabilities.
- 2026-05-12 product_launch Thinking Machines Lab announced its first AI model, focusing on real-time interaction. source
- 2026-05-12 product_launch Thinking Machines Lab announced a research preview of its new "interaction models" designed for real-time, simultaneous AI-human collaboration. source
- 2026-05-12 product_launch Thinking Machines Lab announced its new interaction models enabling simultaneous input processing and response generation.
3 day(s) with sentiment data
Inkling's 1M token context window is a key differentiator for long-context reasoning.
The Inkling model boasts an exceptionally large 1 million token context window, trained on 45 trillion tokens. This is a significant feature that enables advanced reasoning over extended inputs across text, image, and audio. This capability is likely to be a major draw for applications requiring deep understanding of lengthy documents or complex multimodal sequences.
Inkling's 41B active parameters suggest a Mixture-of-Experts (MoE) architecture.
The recent release of Inkling highlights a 975B total parameter count with only 41B active parameters. This significant difference strongly suggests an MoE architecture, where only a subset of parameters are utilized for any given input, enabling efficient scaling. This pattern is consistent with other large models designed for efficiency.
Thinking Machines Lab will release a specialized Inkling fine-tune for enterprise agentic workflows within 90 days.
The integration of Inkling into Databricks' Unity AI Gateway, specifically mentioning optimization for 'coding, agentic reasoning, and multimodal inputs' and connectivity with coding agents, indicates a strong enterprise focus. This suggests Thinking Machines Lab will likely release a version of Inkling specifically fine-tuned for these enterprise agentic workflows to capitalize on this early integration.
Inkling's architecture leverages Mixture-of-Experts (MoE)
The mention of Inkling being a Mixture-of-Experts (MoE) model, particularly with 975B total parameters and 41B active parameters, indicates a sophisticated architecture designed for efficiency and scalability. This is a notable characteristic for an open-weight model.
Inkling's 1M context window is a key differentiator
Multiple sources highlight Inkling's 1 million token context window. This is a significant technical achievement and a potential differentiator in the open-weight model space, especially for applications requiring long-form content understanding or generation.
-
Former Tencent AI Lead Yao Xingcheng Joins Thinking Machines Lab
Yao Xingcheng, the former head of pre-training for Tencent's Hunyuan large language model, has joined the North American startup Thinking Machines Lab. This move comes after Yao Xingcheng left Tencent in July and was re…
-
LLM Fine-Tuning Methods: SFT, LoRA, QLoRA, RFT, and Distillation Explained
The article outlines various methods for fine-tuning large language models, focusing on practical applications and tool choices. Supervised Fine-Tuning (SFT) is presented as a starting point, requiring labeled input-out…
-
Chinese open-source AI models gain traction in US businesses · 1 source tracked
Chinese open-weight AI models are gaining traction with U.S. businesses, challenging the dominance of American companies like OpenAI and Anthropic. These models, such as Moonshot's Kimi K3 and Z.AI's GLM-5.3-Flash, offe…
-
Chinese labs lead open-source AI model frontier, surpassing US releases
The open-source AI model landscape is rapidly evolving, with a notable shift in the frontier of model development. Chinese labs are increasingly releasing larger and more performant models, often surpassing those from A…
-
Thinking Machines Lab releases Inkling-Small open-source MoE model
Thinking Machines Lab has released Inkling-Small, an open-source Mixture of Experts (MoE) model. Despite its smaller size, Inkling-Small surpasses its predecessor in coding tests and offers native multimodality with a 1…
-
Thinking Machines releases Inkling-Small, outperforming larger predecessor
Thinking Machines Lab has launched Inkling-Small, a new open-weights multimodal model that prioritizes efficiency over sheer size. Despite being significantly smaller than its predecessor, Inkling, Inkling-Small demonst…
-
Lilian Weng returns to OpenAI to lead self-evolving AI team
Lilian Weng, a former VP of Research and Safety at OpenAI, has returned to the company to lead its "self-evolving" (RSI) team, a high-priority research initiative. Weng had previously left OpenAI to co-found Thinking Ma…
-
OpenAI's July Revenue Surpasses Q2; Key Researcher Rejoins
OpenAI's Chief Financial Officer, Sarah Friar, announced that the company's annualized revenue in July surpassed its total revenue for the entire second quarter. This growth was discussed during an internal meeting wher…
-
Thinking Machines co-founder Lilian Weng rejoins OpenAI to lead AI research
Lilian Weng, a co-founder of Thinking Machines Lab, has rejoined OpenAI after a brief departure from her startup. Weng cited health reasons for stepping down from Thinking Machines, where she was a co-founder alongside …
-
Mira Murati's Thinking Machines Lab launches Inkling with open weights and honest positioning
Thinking Machines Lab, co-founded by former OpenAI executive Mira Murati, has launched its first model, Inkling. This open-weights system boasts 975 billion parameters and a million-token context window, capable of proc…
-
OpenAI, Moonshot, Anthropic launch flagship models; benchmarks show varied strengths
In a rapid succession of releases, OpenAI, Moonshot AI, and Anthropic have launched their latest flagship models: GPT-5.6 Sol, Kimi K3, and Claude Opus 5, respectively. While all three models offer substantial context w…
-
US AI Labs Launch Open-Source Models to Challenge Chinese Dominance
American AI startups are emerging to challenge the dominance of Chinese open-source models in the AI landscape. Poolside, co-founded by former GitHub CTO Jason Warner, has released a new model named Laguna, which claims…
-
Mira Murati's Inkling model self-fine-tunes on launch day, tops US open-weight charts
Mira Murati's Thinking Machines Lab has released Inkling, a 975B-parameter model that demonstrated self-fine-tuning capabilities on its launch day. The model successfully trained itself to become a lipogram model, avoid…
-
Thinking Machines Lab releases 975B parameter AI model Inkling
Thinking Machines Lab has launched Inkling, an open-weight AI model with 975 billion parameters. The model is inspired by DeepSeek's design and requires 2 TB of GPU memory for operation. Its performance has been evaluat…
-
AI Frontier Rises: Two Labs Drop Major Open-Weight Models in One Day
Two major AI labs, Thinking Machines and Moonshot AI, released significant open-weight models within a single day, signaling a rapid rise in AI capabilities. Thinking Machines launched Inkling, a 975 billion parameter m…
-
Together AI launches Inkling multimodal MoE model with 1M context window
Together AI has launched Inkling, a multimodal Mixture-of-Experts (MoE) model developed by Thinking Machines Lab. This open-weight model boasts 975 billion total parameters with 41 billion active parameters, a 1 million…
-
Big Tech execs warn AI labs are stealing enterprise data
Several prominent tech executives are warning enterprises about the risks of using proprietary AI models, arguing that these labs are effectively extracting valuable business data and know-how. Microsoft CEO Satya Nadel…
-
Thinking Machines Lab emerges as AI competitor to OpenAI
Thinking Machines Lab, an artificial intelligence company, is emerging as a potential competitor to established players like OpenAI. The company is focusing on developing open-weight AI models, positioning itself as a c…
-
Open-weight LLM releases from July 2026 detailed by license and benchmarks
A compilation of open-weight large language model releases from July 2026 categorizes models by their licenses and benchmark performance. Notable releases include Thinking Machines Lab's Inkling under Apache 2.0, DeepSe…
-
Thinking Machines releases Inkling, a 975B parameter open-weights multimodal model
Thinking Machines Lab has released Inkling, a new family of open-weights multimodal foundation models. The flagship Inkling model features 975 billion total parameters with 41 billion active, supports a 1 million token …