All People
Soumith Chintala

Soumith Chintala

PyTorch creator

Recent Activity71 x-posts

Recent Activity

soumithchintala
Soumith Chintala

@soumithchintala

We are scientists, engineers, and builders behind some of the most widely used AI products and libraries, including ChatGPT.

Highlights: Soumith Chintala announces Thinking Machines Lab, highlighting their team's role in creating widely-used AI products like ChatGPT.

Worth reading: It signals a new venture from a key PyTorch co-creator, focusing on building impactful AI products.

InfraLLM
soumithchintala
Soumith Chintala

@soumithchintala

@Mojo_flyin @dwarkesh_sp i'd be excited to have Google and AMD create real ecosystems. I fail to see why Jensen has a burden there. There should be multiple strong American AI ecosystems at the hardware layer. Historically, I struggled to make a dent in conversations around ecosystems with many

Highlights: Chintala advocates for multiple strong American AI hardware ecosystems, including Google and AMD, and dismisses the idea that Jensen (Nvidia) has a burden to support them.

Worth reading: Reveals Chintala's perspective on AI hardware competition and ecosystem diversity.

Infra
soumithchintala
Soumith Chintala

@soumithchintala

Thinky's secret plan: 1: Increase Human<->AI bandwidth 2: Raise ceiling of human+AI intelligence 3: Help humans continue as main-characters in the new world We are at Step 1. Interaction Models are great real-time collaborative tools for humans. Here's a preview:

Highlights: Chintala outlines Thinking Machines Lab's vision: increasing human-AI bandwidth, raising human+AI intelligence ceiling, and keeping humans central. They are at Step 1 with Interaction Models.

Worth reading: Provides insight into the mission of Chintala's new startup, Thinking Machines Lab.

Multi-modalAgent
soumithchintala
Soumith Chintala

@soumithchintala

We are scientists, engineers, and builders behind some of the most widely used AI products and libraries, including ChatGPT.

Highlights: Soumith Chintala announces his involvement with Thinking Machines Lab, highlighting the team's background in building widely used AI products.

Worth reading: It marks his transition from Meta to a new venture, signaling a shift in his career focus.

LLMInfraDeployment
soumithchintala
Soumith Chintala

@soumithchintala

We are scientists, engineers, and builders behind some of the most widely used AI products and libraries, including ChatGPT.

Highlights: Soumith Chintala announces his involvement with Thinking Machines Lab, highlighting the team's background in creating major AI products.

Worth reading: It signals the formation of a new AI startup by key figures from OpenAI and PyTorch.

AgentLLMMulti-modal
soumithchintala
Soumith Chintala

@soumithchintala

Thinky's secret plan: 1: Increase Human<->AI bandwidth 2: Raise ceiling of human+AI intelligence 3: Help humans continue as main-characters in the new world We are at Step 1. Here's a preview:

Highlights: Soumith outlines Thinking Machines' three-step strategy: increasing human-AI bandwidth, raising intelligence ceiling, and keeping humans central.

Worth reading: Reveals the strategic vision behind Thinking Machines Lab's interaction models.

AgentMulti-modalTooling
soumithchintala
Soumith Chintala

@soumithchintala

we've been working on democratizing fast kernel writing on the @PyTorch team. try

Highlights: Soumith Chintala announces efforts to democratize fast kernel writing within the PyTorch team.

Worth reading: Highlights ongoing work to make high-performance kernel development more accessible.

InfraTooling
soumithchintala
Soumith Chintala

@soumithchintala

reading "AI News" (previously Smol Talk) is probably the highest-leverage 45 mins

Highlights: Recommends reading 'AI News' as a high-leverage use of time.

Worth reading: Shows a key figure's trusted information source for AI developments.

LLM
soumithchintala
Soumith Chintala

@soumithchintala

MacStudio you ask? Apple Engineering's **actual** time spent on PyTorch support

Highlights: Highlights Apple Engineering's investment in PyTorch support.

Worth reading: Indicates Apple's commitment to PyTorch, relevant for ML practitioners.

Infra
soumithchintala
Soumith Chintala

@soumithchintala

Sometimes we forget that NVIDIA wins because it's a software company.

Highlights: Argues that NVIDIA's success is due to its software, not just hardware.

Worth reading: Provides insight into a key competitive advantage in AI hardware.

Infra
soumithchintala
Soumith Chintala

@soumithchintala

ChatGPT seems to be **really** good for creative work and a solid starting point

Highlights: Praises ChatGPT for creative tasks and as a starting point.

Worth reading: Reflects early positive reception of ChatGPT from a prominent AI figure.

LLM
soumithchintala
Soumith Chintala

@soumithchintala

MacStudio you ask? Apple Engineering's **actual** time spent on PyTorch support

Highlights: Comments on Apple Engineering's time spent on PyTorch support, likely in response to a question about MacStudio.

Worth reading: Provides insight into Apple's engineering priorities regarding PyTorch.

Infra
soumithchintala
Soumith Chintala

@soumithchintala

Sometimes we forget that NVIDIA wins because it's a software company.

Highlights: Points out that NVIDIA's success is due to its software, not just hardware.

Worth reading: Offers a perspective on NVIDIA's competitive advantage.

Infra
soumithchintala
Soumith Chintala

@soumithchintala

reading "AI News" (previously Smol Talk) is probably the highest-leverage 45 mins

Highlights: Recommends reading 'AI News' as a high-leverage activity.

Worth reading: Shows what the PyTorch co-founder considers valuable for staying informed.

LLMInfra
soumithchintala
Soumith Chintala

@soumithchintala

Sometimes we forget that NVIDIA wins because it's a software company.

Highlights: Attributes NVIDIA's success to its software, not just hardware.

Worth reading: Provides insight into the competitive advantage of a key AI hardware company.

InfraDeployment
soumithchintala
Soumith Chintala

@soumithchintala

anyone else feel burned out by a new AI breakthrough every week?

Highlights: Expresses fatigue from the rapid pace of AI breakthroughs.

Worth reading: Reflects a common sentiment in the AI community about news overload.

LLM
soumithchintala
Soumith Chintala

@soumithchintala

reading 'AI News' (previously Smol Talk) is probably the highest-leverage 45 mins

Highlights: Soumith recommends the AI News newsletter as a high-leverage read.

Worth reading: Shows his endorsement of a resource for staying updated on AI.

LLMEvaluation
soumithchintala
Soumith Chintala

@soumithchintala

Sometimes we forget that NVIDIA wins because it's a software company.

Highlights: He emphasizes NVIDIA's success is due to software, not just hardware.

Worth reading: Insight into the strategic advantage of software in AI hardware.

InfraDeployment
soumithchintala
Soumith Chintala

@soumithchintala

Open LLMs need to get organized and co-ordinated about sharing human feedback.

Highlights: He calls for better coordination among open LLM developers on human feedback sharing.

Worth reading: Highlights a key challenge for open-source AI development.

LLMFine-tuningSafety
soumithchintala
Soumith Chintala

@soumithchintala

reading "AI News" (previously Smol Talk) is probably the highest-leverage 45 mins

Highlights: Soumith recommends reading 'AI News' as a high-leverage activity.

Worth reading: Shows what a leading AI figure considers valuable for staying informed.

LLM
soumithchintala
Soumith Chintala

@soumithchintala

Sometimes we forget that NVIDIA wins because it's a software company.

Highlights: Soumith argues NVIDIA's success is due to its software, not just hardware.

Worth reading: Insightful take on the importance of software ecosystem in AI hardware dominance.

Infra
soumithchintala
Soumith Chintala

@soumithchintala

MacStudio you ask? Apple Engineering's **actual** time spent on PyTorch support

Highlights: Comments on Apple's engineering effort for PyTorch on MacStudio.

Worth reading: Reveals platform support dynamics for PyTorch.

Infra
soumithchintala
Soumith Chintala

@soumithchintala

MacStudio you ask? Apple Engineering's **actual** time spent on PyTorch support

Highlights: He comments on Apple's engineering effort for PyTorch support on MacStudio.

Worth reading: Insight into hardware-software collaboration for AI.

InfraDeployment
soumithchintala
Soumith Chintala

@soumithchintala

Sometimes we forget that NVIDIA wins because it's a software company.

Highlights: He emphasizes NVIDIA's software strength as key to its success.

Worth reading: Highlights the importance of software in AI hardware dominance.

InfraDeployment
soumithchintala
Soumith Chintala

@soumithchintala

MacStudio you ask? Apple Engineering's **actual** time spent on PyTorch support

Highlights: Comments on Apple Engineering's time investment in PyTorch support.

Worth reading: Reveals industry dynamics around hardware-software collaboration.

Infra
soumithchintala
Soumith Chintala

@soumithchintala

Sometimes we forget that NVIDIA wins because it's a software company.

Highlights: Attributes NVIDIA's success to software, not just hardware.

Worth reading: Insight into competitive advantage in AI infrastructure.

Infra
soumithchintala
Soumith Chintala

@soumithchintala

Sometimes we forget that NVIDIA wins because it's a software company.

Highlights: Argues that NVIDIA's success is due to its software, not just hardware.

Worth reading: Offers insight into NVIDIA's competitive advantage in AI.

Infra
soumithchintala
Soumith Chintala

@soumithchintala

MacStudio you ask? Apple Engineering's **actual** time spent on PyTorch support

Highlights: Soumith comments on Apple Engineering's actual time spent on PyTorch support, likely in response to a question about MacStudio.

Worth reading: Reveals the behind-the-scenes effort for PyTorch on Apple hardware.

InfraTooling
soumithchintala
Soumith Chintala

@soumithchintala

Sometimes we forget that NVIDIA wins because it's a software company.

Highlights: Soumith highlights NVIDIA's success is due to its software, not just hardware.

Worth reading: Provides insight into the strategic advantage of NVIDIA in AI.

Infra
soumithchintala
Soumith Chintala

@soumithchintala

MacStudio you ask? Apple Engineering's **actual** time spent on PyTorch support

Highlights: Comments on Apple's engineering effort for PyTorch on MacStudio.

Worth reading: Reveals behind-the-scenes of PyTorch hardware support.

InfraDeployment
soumithchintala
Soumith Chintala

@soumithchintala

Sometimes we forget that NVIDIA wins because it's a software company.

Highlights: He emphasizes NVIDIA's software strength.

Worth reading: Insight into competitive advantage in AI hardware.

Infra
soumithchintala
Soumith Chintala

@soumithchintala

MacStudio you ask? Apple Engineering's **actual** time spent on PyTorch support

Highlights: Comments on Apple's engineering effort for PyTorch support on Mac Studio.

Worth reading: Insight into hardware-software collaboration for AI frameworks.

InfraDeployment
soumithchintala
Soumith Chintala

@soumithchintala

MacStudio you ask? Apple Engineering's **actual** time spent on PyTorch support

Highlights: Comments on Apple's engineering effort for PyTorch on Mac Studio.

Worth reading: Reveals behind-the-scenes of PyTorch hardware support.

Infra
soumithchintala
Soumith Chintala

@soumithchintala

Sometimes we forget that NVIDIA wins because it's a software company.

Highlights: Attributes NVIDIA's success to software, not just hardware.

Worth reading: Insight into the strategic advantage of software in AI hardware companies.

Infra
soumithchintala
Soumith Chintala

@soumithchintala

we've been working on democratizing fast kernel writing on the @PyTorch team. try

Highlights: Soumith highlights efforts to democratize fast kernel writing within the PyTorch team.

Worth reading: Shows ongoing commitment to making PyTorch more accessible and efficient.

Infra
soumithchintala
Soumith Chintala

@soumithchintala

we've been working on democratizing fast kernel writing on the @PyTorch team. try

Highlights: Soumith announces efforts to democratize fast kernel writing in PyTorch.

Worth reading: Highlights PyTorch's focus on making high-performance kernel development accessible.

InfraTooling
soumithchintala
Soumith Chintala

@soumithchintala

we've been working on democratizing fast kernel writing on the @PyTorch team. try

Highlights: Announces work on democratizing fast kernel writing in PyTorch.

Worth reading: Highlights PyTorch's push to make kernel development more accessible.

InfraTooling
soumithchintala
Soumith Chintala

@soumithchintala

we've been working on democratizing fast kernel writing on the @PyTorch team. try

Highlights: Soumith highlights PyTorch team's efforts to make fast kernel writing more accessible.

Worth reading: Indicates ongoing work to lower barriers for high-performance ML code.

InfraDeployment
soumithchintala
Soumith Chintala

@soumithchintala

we've been working on democratizing fast kernel writing on the @PyTorch team. try

Highlights: Soumith announces PyTorch team's effort to democratize fast kernel writing.

Worth reading: Highlights ongoing work to make high-performance kernel development accessible.

InfraTooling
soumithchintala
Soumith Chintala

@soumithchintala

Today, we are excited to announce Thinking Machines Lab (thinkingmachines.ai), an artificial intelligence research and product company.

Highlights: Soumith announces the launch of Thinking Machines Lab, an AI research and product company.

Worth reading: Announces a major new venture from a PyTorch co-founder.

Infra
soumithchintala
Soumith Chintala

@soumithchintala

I'm giving the opening Keynote at ICML 2024 on Tuesday the 23rd @ 9:30am CEST.

Highlights: Announcing a keynote at ICML 2024.

Worth reading: Highlights his role in the AI research community.

LLM
soumithchintala
Soumith Chintala

@soumithchintala

I'm giving the opening Keynote at ICML 2024 on Tuesday the 23rd @ 9:30am CEST.

Highlights: Soumith announced his opening keynote at ICML 2024.

Worth reading: Highlights his prominence in the ML community.

LLM
soumithchintala
Soumith Chintala

@soumithchintala

Sometimes we forget that NVIDIA wins because it's a software company.

Highlights: Soumith points out that NVIDIA's success is due to its software, not just hardware.

Worth reading: Provides insight into the strategic advantage of NVIDIA in AI.

Infra
soumithchintala
Soumith Chintala

@soumithchintala

MacStudio you ask? Apple Engineering's **actual** time spent on PyTorch support

Highlights: Comments on Apple Engineering's effort on PyTorch support.

Worth reading: Insight into industry collaboration between Apple and PyTorch.

InfraDeployment
soumithchintala
Soumith Chintala

@soumithchintala

reading "AI News" (previously Smol Talk) is probably the highest-leverage 45 mins

Highlights: Soumith recommends reading 'AI News' as a high-leverage use of time.

Worth reading: Shows what an AI leader considers valuable for staying informed.

LLM
soumithchintala
Soumith Chintala

@soumithchintala

reading "AI News" (previously Smol Talk) is probably the highest-leverage 45 mins

Highlights: Recommends reading 'AI News' as a high-leverage use of time.

Worth reading: Shows where Soumith finds value in AI news consumption.

LLM
soumithchintala
Soumith Chintala

@soumithchintala

reading "AI News" (previously Smol Talk) is probably the highest-leverage 45 mins

Highlights: Soumith recommends reading 'AI News' as a highly efficient way to stay informed.

Worth reading: Shows his perspective on maximizing learning in AI.

LLMInfra
soumithchintala
Soumith Chintala

@soumithchintala

reading "AI News" (previously Smol Talk) is probably the highest-leverage 45 mins

Highlights: Soumith recommends reading AI News as a high-leverage activity for 45 minutes.

Worth reading: Shows where a leading AI figure focuses his learning time.

LLM
soumithchintala
Soumith Chintala

@soumithchintala

reading "AI News" (previously Smol Talk) is probably the highest-leverage 45 mins

Highlights: Recommends a newsletter as high-leverage reading.

Worth reading: Shows what a key figure in AI values for staying informed.

LLM
soumithchintala
Soumith Chintala

@soumithchintala

reading "AI News" (previously Smol Talk) is probably the highest-leverage 45 mins

Highlights: Recommends a newsletter as high-leverage reading.

Worth reading: Shows what an AI leader finds valuable.

LLM
soumithchintala
Soumith Chintala

@soumithchintala

reading "AI News" (previously Smol Talk) is probably the highest-leverage 45 mins

Highlights: Soumith recommends reading 'AI News' as a high-leverage activity.

Worth reading: Shows what a key AI figure considers valuable for staying informed.

LLM
soumithchintala
Soumith Chintala

@soumithchintala

reading "AI News" (previously Smol Talk) is probably the highest-leverage 45 mins

Highlights: Soumith recommends reading 'AI News' as a high-leverage activity.

Worth reading: Insight into what Soumith values for staying informed in AI.

LLM
soumithchintala
Soumith Chintala

@soumithchintala

reading 'AI News' (previously Smol Talk) is probably the highest-leverage 45 mins

Highlights: Soumith recommends reading 'AI News' as a high-leverage activity.

Worth reading: Shows what a leading AI figure considers valuable for staying informed.

LLM
soumithchintala
Soumith Chintala

@soumithchintala

reading "AI News" (previously Smol Talk) is probably the highest-leverage 45 mins

Highlights: Recommends reading AI News as a high-leverage activity.

Worth reading: Shows where a top AI researcher spends their time for maximum impact.

LLMDeployment
soumithchintala
Soumith Chintala

@soumithchintala

reading "AI News" (previously Smol Talk) is probably the highest-leverage 45 mins

Highlights: Recommends a newsletter called AI News as a high-leverage way to stay informed.

Worth reading: Shows Chintala's perspective on efficient learning in AI.

LLM
soumithchintala
Soumith Chintala

@soumithchintala

reading "AI News" (previously Smol Talk) is probably the highest-leverage 45 mins

Highlights: Recommends reading 'AI News' as a high-leverage activity for staying informed.

Worth reading: Shows a key AI figure's perspective on efficient learning.

LLM
soumithchintala
Soumith Chintala

@soumithchintala

reading "AI News" (previously Smol Talk) is probably the highest-leverage 45 mins

Highlights: Recommends a 45-minute AI news digest as high-leverage activity.

Worth reading: Shows where a top AI figure invests his learning time.

LLM
soumithchintala
Soumith Chintala

@soumithchintala

reading "AI News" (previously Smol Talk) is probably the highest-leverage 45 mins

Highlights: Soumith recommends reading 'AI News' as a high-leverage activity.

Worth reading: Shows what a top AI researcher considers valuable for staying informed.

LLM
soumithchintala
Soumith Chintala

@soumithchintala

reading "AI News" (previously Smol Talk) is probably the highest-leverage 45 mins

Highlights: Soumith recommends reading 'AI News' as a high-leverage use of time.

Worth reading: Shows his perspective on efficient learning in AI.

LLM
soumithchintala
Soumith Chintala

@soumithchintala

reading 'AI News' (previously Smol Talk) is probably the highest-leverage 45 mins

Highlights: Soumith recommends 'AI News' as a high-leverage use of 45 minutes.

Worth reading: Shows what the PyTorch founder considers valuable for staying informed.

LLMTooling
soumithchintala
Soumith Chintala

@soumithchintala

reading "AI News" (previously Smol Talk) is probably the highest-leverage 45 mins

Highlights: Soumith recommends reading 'AI News' as a high-leverage use of time.

Worth reading: Shows what an AI leader considers valuable for staying informed.

LLM
soumithchintala
Soumith Chintala

@soumithchintala

reading "AI News" (previously Smol Talk) is probably the highest-leverage 45 mins

Highlights: Soumith recommends a newsletter as high-leverage reading.

Worth reading: Shows his opinion on efficient learning resources.

LLM
soumithchintala
Soumith Chintala

@soumithchintala

Open LLMs need to get organized and co-ordinated about sharing human feedback.

Highlights: Calls for open LLM community to coordinate on sharing human feedback.

Worth reading: Highlights a key challenge in open-source LLM development.

LLMFine-tuning
soumithchintala
Soumith Chintala

@soumithchintala

Open LLMs need to get organized and co-ordinated about sharing human feedback.

Highlights: He advocates for better coordination among open LLM projects regarding human feedback.

Worth reading: Reflects his views on open-source AI collaboration.

LLMFine-tuning
soumithchintala
Soumith Chintala

@soumithchintala

Open LLMs need to get organized and co-ordinated about sharing human feedback.

Highlights: Soumith calls for better coordination among open LLM projects on human feedback sharing.

Worth reading: Addresses a key challenge in open-source LLM development.

LLMFine-tuning
soumithchintala
Soumith Chintala

@soumithchintala

Open LLMs need to get organized and co-ordinated about sharing human feedback.

Highlights: Calls for coordination among open LLM developers on human feedback.

Worth reading: Highlights a key challenge for open-source AI alignment.

LLMSafety
soumithchintala
Soumith Chintala

@soumithchintala

Open LLMs need to get organized and co-ordinated about sharing human feedback.

Highlights: Calls for coordination in open LLM community on feedback sharing.

Worth reading: Highlights a key challenge for open-source LLMs.

LLMFine-tuning
soumithchintala
Soumith Chintala

@soumithchintala

Open LLMs need to get organized and co-ordinated about sharing human feedback.

Highlights: Advocates for coordination among open LLM projects to share human feedback data.

Worth reading: Key call to action for open-source AI community collaboration.

Fine-tuningLLM
soumithchintala
Soumith Chintala

@soumithchintala

ChatGPT seems to be **really** good for creative work and a solid starting point

Highlights: Acknowledges ChatGPT's strength in creative tasks.

Worth reading: Early endorsement of ChatGPT's utility from a key AI figure.

LLMMulti-modal
soumithchintala
Soumith Chintala

@soumithchintala

anyone else feel burned out by a new AI breakthrough every week?

Highlights: Expresses burnout from the rapid pace of AI breakthroughs.

Worth reading: Reflects the sentiment of many in the AI community.

SafetyEvaluation
soumithchintala
Soumith Chintala

@soumithchintala

anyone else feel burned out by a new AI breakthrough every week?

Highlights: Expresses fatigue from rapid AI progress pace.

Worth reading: Reflects sentiment in AI community about news overload.

LLM
71 x-posts · All time