Alexander Jackson (@AI_Workflow_Hub) header
Alexander Jackson (@AI_Workflow_Hub) profile picture

AI tips, tricks & daily automation 🤖 Smart systems for faster growth 📩 DM for partnerships & collaborations [email protected]

New York, USA
Joined March 2011
Photo by Alexander Jackson (@AI_Workflow_Hub)Photo by Alexander Jackson (@AI_Workflow_Hub)Photo by Alexander Jackson (@AI_Workflow_Hub)Photo by Alexander Jackson (@AI_Workflow_Hub)Photo by Alexander Jackson (@AI_Workflow_Hub)Photo by Alexander Jackson (@AI_Workflow_Hub)Photo by Alexander Jackson (@AI_Workflow_Hub)Photo by Alexander Jackson (@AI_Workflow_Hub)Photo by Alexander Jackson (@AI_Workflow_Hub)Photo by Alexander Jackson (@AI_Workflow_Hub)
Every new AI video model promises cinematic quality, but consistency is what really matters. This was created with Seedance 2.5 on Morphic, and the motion, camera work, and visual quality stayed impressive from start to finish. If this is where AI video is heading, creators are about to unlock a whole new level of storytelling. @morphic Morphic.com
83
113
161
97,215
Alexander Jackson retweeted
Who knew black and white pixels could make such a difference? I tried the same 1-bit Dither prompt on two image models, and I love how differently they interpreted this retro aesthetic. Kling Image 4.0 went for that classic pixel-art feel: bold black-and-white shapes, crisp edges, and beautifully consistent pixel patterns. It even keeps the original subject instantly recognizable. GPT Image 2.6 brought a more textured, grainy interpretation, giving the image a different kind of artistic character. Personally, I'm drawn to Kling's clean, authentic 1-bit look. Every little pixel just clicks into place. Kling's limited-time Black Friday offer: Get 2K images for as low as $0.022/image, and native 8K for as low as $0.033/image! Swipe to compare! 👉 Image 1: Generated with Kling Image 4.0 👉 Image 2: Generated with GPT Image 2.6
19
38
43
99,652
Alexander Jackson retweeted
Medical AI is getting seriously powerful. LingSante-3.1-air-VL can understand medical images, generate reports, search medical literature, and even interact with healthcare apps. All powered by a 124B parameter model with just 5.5B active parameters. Open-source multimodal AI is entering a whole new territory.
Meet LingSante-3.1-air-VL, an open-source multimodal medical model with 124B total and 5.5B active parameters. It combines medical image understanding, report generation and medical reasoning with multimodal search and tool use. In our evaluation, it leads the compared models on AFUMED-Derm, AFUMED-Report, MIMIC-CXR and medical VQA (averaged across benchmarks), ahead of GPT-6.1 Sol, Gemini 3.8 Flash on all four. It also demonstrates strong general-purpose capabilities, scoring higher than GPT-6.1 Sol on BrowseComp-VL and Gemini 3.8 Flash on AndroidWorld, alongside competitive performance on MMSearch. Explore LingSante-3.1-air-VL and build with it.
6
8
7
278
Alexander Jackson retweeted
Medical Al is becoming increasingly sophisticated. LingSante-3.1-air-VL combines image understanding, report generation, medical reasoning, and evidence-guided search. Its ability to connect different healthcare tasks makes it more interesting than another model focused on a single benchmark.
Meet LingSante-3.1-air-VL, an open-source multimodal medical model with 124B total and 5.5B active parameters. It combines medical image understanding, report generation and medical reasoning with multimodal search and tool use. In our evaluation, it leads the compared models on AFUMED-Derm, AFUMED-Report, MIMIC-CXR and medical VQA (averaged across benchmarks), ahead of GPT-6.1 Sol, Gemini 3.8 Flash on all four. It also demonstrates strong general-purpose capabilities, scoring higher than GPT-6.1 Sol on BrowseComp-VL and Gemini 3.8 Flash on AndroidWorld, alongside competitive performance on MMSearch. Explore LingSante-3.1-air-VL and build with it.
7
8
10
311
Alexander Jackson retweeted
Medical AI needs accuracy, context, and evidence. LingSante-3.1-air-VL focuses on all three. It connects medical images with diagnostic reasoning, searches relevant literature, and evaluates whether the available evidence is sufficient. That’s a meaningful direction for building more capable medical AI systems.
Meet LingSante-3.1-air-VL, an open-source multimodal medical model with 124B total and 5.5B active parameters. It combines medical image understanding, report generation and medical reasoning with multimodal search and tool use. In our evaluation, it leads the compared models on AFUMED-Derm, AFUMED-Report, MIMIC-CXR and medical VQA (averaged across benchmarks), ahead of GPT-6.1 Sol, Gemini 3.8 Flash on all four. It also demonstrates strong general-purpose capabilities, scoring higher than GPT-6.1 Sol on BrowseComp-VL and Gemini 3.8 Flash on AndroidWorld, alongside competitive performance on MMSearch. Explore LingSante-3.1-air-VL and build with it.
1
2
2
134
Alexander Jackson retweeted
A new future begins with one spark of imagination. ⚡🌆 The future is not waiting for us. We’re creating it with AI. 🚀 #AI #AIVideo #FutureTech #ArtificialIntelligence #DigitalInnovation #AICreativity #Tech
🤖 Made with AI
8
9
223
Alexander Jackson retweeted
College degrees AI will make WORTHLESS by 2030 🔴 ALREADY DEAD ・Journalism ・Graphic design ・Business admin 🟠 LOSING VALUE FAST ・Computer science ・Finance ・Marketing 🟢 THE ONES WORTH MORE THAN EVER ↓
1
2
3
59
Alexander Jackson retweeted
Imagine uploading a medical image and getting more than a basic description. LingSante-3.1-air-VL can analyze the image, connect it with context, generate a report, and search for supporting evidence. This is the kind of multimodal workflow worth paying attention to.
Meet LingSante-3.1-air-VL, an open-source multimodal medical model with 124B total and 5.5B active parameters. It combines medical image understanding, report generation and medical reasoning with multimodal search and tool use. In our evaluation, it leads the compared models on AFUMED-Derm, AFUMED-Report, MIMIC-CXR and medical VQA (averaged across benchmarks), ahead of GPT-6.1 Sol, Gemini 3.8 Flash on all four. It also demonstrates strong general-purpose capabilities, scoring higher than GPT-6.1 Sol on BrowseComp-VL and Gemini 3.8 Flash on AndroidWorld, alongside competitive performance on MMSearch. Explore LingSante-3.1-air-VL and build with it.
4
8
8
104
Alexander Jackson retweeted
A 124B parameter medical AI model with only 5.5B active parameters. That's already interesting. But LingSante-3.1-air-VL also brings image understanding, medical reasoning, evidence search, and tool interaction together. Open-source multimodal AI keeps raising the bar.
Meet LingSante-3.1-air-VL, an open-source multimodal medical model with 124B total and 5.5B active parameters. It combines medical image understanding, report generation and medical reasoning with multimodal search and tool use. In our evaluation, it leads the compared models on AFUMED-Derm, AFUMED-Report, MIMIC-CXR and medical VQA (averaged across benchmarks), ahead of GPT-6.1 Sol, Gemini 3.8 Flash on all four. It also demonstrates strong general-purpose capabilities, scoring higher than GPT-6.1 Sol on BrowseComp-VL and Gemini 3.8 Flash on AndroidWorld, alongside competitive performance on MMSearch. Explore LingSante-3.1-air-VL and build with it.
5
9
9
238
Alexander Jackson retweeted
What makes medical AI useful isn't just recognizing what's in an image. It's understanding why it matters. LingSante-3.1-air-VL combines medical vision, reasoning, report generation, and evidence search to move beyond simple image descriptions. That's a meaningful improvement in direction.
Meet LingSante-3.1-air-VL, an open-source multimodal medical model with 124B total and 5.5B active parameters. It combines medical image understanding, report generation and medical reasoning with multimodal search and tool use. In our evaluation, it leads the compared models on AFUMED-Derm, AFUMED-Report, MIMIC-CXR and medical VQA (averaged across benchmarks), ahead of GPT-6.1 Sol, Gemini 3.8 Flash on all four. It also demonstrates strong general-purpose capabilities, scoring higher than GPT-6.1 Sol on BrowseComp-VL and Gemini 3.8 Flash on AndroidWorld, alongside competitive performance on MMSearch. Explore LingSante-3.1-air-VL and build with it.
12
15
15
117
Alexander Jackson retweeted
Medical AI is becoming more than a chatbot. LingSante-3.1-air-VL can analyze images, generate clinical-style reports, search medical literature, and navigate healthcare applications. The ability to combine multiple steps into one workflow makes this release particularly interesting.
Meet LingSante-3.1-air-VL, an open-source multimodal medical model with 124B total and 5.5B active parameters. It combines medical image understanding, report generation and medical reasoning with multimodal search and tool use. In our evaluation, it leads the compared models on AFUMED-Derm, AFUMED-Report, MIMIC-CXR and medical VQA (averaged across benchmarks), ahead of GPT-6.1 Sol, Gemini 3.8 Flash on all four. It also demonstrates strong general-purpose capabilities, scoring higher than GPT-6.1 Sol on BrowseComp-VL and Gemini 3.8 Flash on AndroidWorld, alongside competitive performance on MMSearch. Explore LingSante-3.1-air-VL and build with it.
4
7
9
368
Alexander Jackson retweeted
Medical AI is entering an exciting phase. LingSante-3.1-air-VL combines medical image understanding, structured report generation, multimodal search, and healthcare app navigation. Instead of handling one task at a time, it’s designed to connect multiple steps into a single workflow.
Meet LingSante-3.1-air-VL, an open-source multimodal medical model with 124B total and 5.5B active parameters. It combines medical image understanding, report generation and medical reasoning with multimodal search and tool use. In our evaluation, it leads the compared models on AFUMED-Derm, AFUMED-Report, MIMIC-CXR and medical VQA (averaged across benchmarks), ahead of GPT-6.1 Sol, Gemini 3.8 Flash on all four. It also demonstrates strong general-purpose capabilities, scoring higher than GPT-6.1 Sol on BrowseComp-VL and Gemini 3.8 Flash on AndroidWorld, alongside competitive performance on MMSearch. Explore LingSante-3.1-air-VL and build with it.
1
3
3
66
Alexander Jackson retweeted
This is where healthcare AI gets exciting. LingSante-3.1-flash can work through complex medical scenarios, retrieve relevant evidence, and execute multi-step tasks. It's not just another model with impressive numbers. The real focus is practical capability.
Today, we release LingSante-3.1-flash, an open-weight MoE model with 560B total parameters and 25B activations per token. It unifies diverse medical capabilities while retaining general capabilities, including complex medical reasoning, long-horizon workflow execution, and professional behavior alignment. LingSante-3.1-flash moves beyond medical QA toward real-world healthcare problem. The model leads on HealthBench Professional (65.72), HLE (text-only, with tools: 72.22) and BrowseComp (with context: 94.10), outperforming GPT-6 Astra, Claude Opus 5.5 and Gemini 4 Argon on all three. LingSante-3.1-flash also performs well on professional long-horizon medical workflows, such as PhysicianBench (57) and EHR-Complex (74.40). Together, these results establish LingSante-3.1-flash as the leading open-weight models for healthcare. The weights are available. Explore LingSante and build with it. More below ↓
6
7
7
366
Alexander Jackson retweeted
AI that understands medical images is impressive. AI that can also connect those images with patient context, medical literature, and structured reporting is even more interesting. LingSante-3.1-air-VL is pushing multimodal medical intelligence in that direction.
Meet LingSante-3.1-air-VL, an open-source multimodal medical model with 124B total and 5.5B active parameters. It combines medical image understanding, report generation and medical reasoning with multimodal search and tool use. In our evaluation, it leads the compared models on AFUMED-Derm, AFUMED-Report, MIMIC-CXR and medical VQA (averaged across benchmarks), ahead of GPT-6.1 Sol, Gemini 3.8 Flash on all four. It also demonstrates strong general-purpose capabilities, scoring higher than GPT-6.1 Sol on BrowseComp-VL and Gemini 3.8 Flash on AndroidWorld, alongside competitive performance on MMSearch. Explore LingSante-3.1-air-VL and build with it.
1
3
3
163
Alexander Jackson retweeted
🧵 AI could cut the time needed to discover new medicines by up to 50%. This is no longer just a prediction. Some of the world’s biggest pharmaceutical companies are already using AI to design molecules, identify promising drug candidates, and detect potential safety risks. Here’s what’s happening. 👇
4
7
9
275
Alexander Jackson retweeted
From pixels to medical insights. LingSante-3.1-air-VL connects medical images with reasoning, generates structured reports, and searches for evidence to support its analysis. It's exciting to see open-source models tackling problems that require more than basic visual understanding.
Meet LingSante-3.1-air-VL, an open-source multimodal medical model with 124B total and 5.5B active parameters. It combines medical image understanding, report generation and medical reasoning with multimodal search and tool use. In our evaluation, it leads the compared models on AFUMED-Derm, AFUMED-Report, MIMIC-CXR and medical VQA (averaged across benchmarks), ahead of GPT-6.1 Sol, Gemini 3.8 Flash on all four. It also demonstrates strong general-purpose capabilities, scoring higher than GPT-6.1 Sol on BrowseComp-VL and Gemini 3.8 Flash on AndroidWorld, alongside competitive performance on MMSearch. Explore LingSante-3.1-air-VL and build with it.
7
9
10
272
Alexander Jackson retweeted
The creativity behind this AI video is incredible Generated with Seedance 2.5 Every frame tells a story, but the real magic is in how the tension, emotion, and action come together. 🎬⚡ This is the kind of AI cinematic storytelling that makes you forget you're watching a generated video. From dramatic shots to character expressions, the details make all the difference. AI filmmaking is getting seriously impressive. 🔥
25
71
121
3,835
Alexander Jackson retweeted
YOUR COMPANY MAY HAVE AI AGENTS NOBODY KNOWS ABOUT A forgotten AI agent can become a security risk. It may still have access to business data, internal systems, or sensitive credentials—even after the employee who created it has moved on. A 2026 Cloud Security Alliance survey revealed a surprising problem. 82% of surveyed organizations had discovered unknown AI agents in their environments. Here's why that matters. 🧵👇
30
103
137
3,155
Alexander Jackson retweeted
Healthcare AI is no longer just about asking questions and getting answers. LingSante-3.1-flash brings together medical reasoning, multi-agent research, and long-horizon workflows. This is a much more practical direction for building capable medical AI systems.
Today, we release LingSante-3.1-flash, an open-weight MoE model with 560B total parameters and 25B activations per token. It unifies diverse medical capabilities while retaining general capabilities, including complex medical reasoning, long-horizon workflow execution, and professional behavior alignment. LingSante-3.1-flash moves beyond medical QA toward real-world healthcare problem. The model leads on HealthBench Professional (65.72), HLE (text-only, with tools: 72.22) and BrowseComp (with context: 94.10), outperforming GPT-6 Astra, Claude Opus 5.5 and Gemini 4 Argon on all three. LingSante-3.1-flash also performs well on professional long-horizon medical workflows, such as PhysicianBench (57) and EHR-Complex (74.40). Together, these results establish LingSante-3.1-flash as the leading open-weight models for healthcare. The weights are available. Explore LingSante and build with it. More below ↓
8
14
15
227
Alexander Jackson retweeted
This launch shows how far specialized AI models are evolving. LingSante-3.1-flash combines 560B total parameters with advanced medical reasoning and workflow execution. Open weights make it even more exciting for developers building the next generation of healthcare tools.
Today, we release LingSante-3.1-flash, an open-weight MoE model with 560B total parameters and 25B activations per token. It unifies diverse medical capabilities while retaining general capabilities, including complex medical reasoning, long-horizon workflow execution, and professional behavior alignment. LingSante-3.1-flash moves beyond medical QA toward real-world healthcare problem. The model leads on HealthBench Professional (65.72), HLE (text-only, with tools: 72.22) and BrowseComp (with context: 94.10), outperforming GPT-6 Astra, Claude Opus 5.5 and Gemini 4 Argon on all three. LingSante-3.1-flash also performs well on professional long-horizon medical workflows, such as PhysicianBench (57) and EHR-Complex (74.40). Together, these results establish LingSante-3.1-flash as the leading open-weight models for healthcare. The weights are available. Explore LingSante and build with it. More below ↓
16
25
27
561
Medical AI is becoming more than a chatbot. LingSante-3.1-air-VL can analyze images, generate clinical-style reports, search medical literature, and navigate healthcare applications. The ability to combine multiple steps into one workflow makes this release particularly interesting.
Meet LingSante-3.1-air-VL, an open-source multimodal medical model with 124B total and 5.5B active parameters. It combines medical image understanding, report generation and medical reasoning with multimodal search and tool use. In our evaluation, it leads the compared models on AFUMED-Derm, AFUMED-Report, MIMIC-CXR and medical VQA (averaged across benchmarks), ahead of GPT-6.1 Sol, Gemini 3.8 Flash on all four. It also demonstrates strong general-purpose capabilities, scoring higher than GPT-6.1 Sol on BrowseComp-VL and Gemini 3.8 Flash on AndroidWorld, alongside competitive performance on MMSearch. Explore LingSante-3.1-air-VL and build with it.
4
7
9
368