AI Tech | Content Creator | Software Developer Engineering | Dm For Collab
Austin, TX
Joined December 2020
- Tweets58.9K
- Following4.7K
- Followers30.9K
- Likes47.6K
Anderson retweeted
T1 going for four in a row. Gen.G in as the No. 1 seed. HLE holding MSI.
And a robotics data company bought a box at the final so 35 engineers who are cracked at League can watch it with the Mecka AI team.
This is crazyyy 😭 Who's taking it?
🤝 Paid partnership · 🤖 Made with AI
The league of legends championship is in the usa this year.. and we booked the largest private box in the stadium
@MeckaAI cofounder @jasontheutopian and first engineer Anthony, both ranked challenger as top 50 players in the world
So we’re inviting 30 elite engineers, researchers and operators to join us in the ownership suite to watch the finals front and center
Anyone with world-class skill in a video game, not just in league, send us your rank or evidence of exceptional ability in what you do
The championship is nov 14 at the barclays center in new york
Apply by Nov 8. Link in comments
Anderson retweeted
OCR’s real headache isn’t reading pages it’s catching mistakes after they hit your workflow.
MiniCPM-OCR pairs a fast 1B OCR model with a separate quality-checker, making document AI more trustworthy.
Your OCR says it’s done—but did it get everything right? 🤔 In high-stakes workflows, even a small error can be costly.
🚀 Meet the MiniCPM-OCR suite: an open-source solution to parse fast and guard the results.
Two new models:
🔹 MiniCPM-OCR 1.0 — SOTA OCR performance and speed at 1B scale. Compact, accurate, and fast.
🔹 MiniCPM-OCR Jev — A 2B decision model that helps assess parsing accuracy, with performance approaching that of Qwen3.5-397B-A17B on JevOCRBench.
⚡ What’s new?
Small model, SOTA results: MiniCPM-OCR 1.0 scores 93.68 on Real5-OmniDocBench and 54.7 on OCRBench v2.
A sharp goalkeeper 🛡️: MiniCPM-OCR Jev is model-agnostic, checking text, formulas, and tables from different OCR parsers. It achieves 93.6 on JevOCRBench, outperforming JEV-35B-A3B (90.1) and approaching Qwen3.5-397B-A17B (93.8).
Up to 128 pages per minute 🤯: With DSpark, MiniCPM-OCR runs at 1.49× the speed of PaddleOCR-VL-1.6 and 3.62× that of MinerU2.5 Pro. The SDK’s concurrency optimizations further boost throughput to as much as 128 pages per minute.
Deploy where you work: Day-0 support for Arm, Houmo AI, Intel, and NVIDIA, plus 10 more chip vendors through FlagOS.
Ready for your stack: vLLM, SGLang, llama.cpp, and Ollama, with quantized models, fine-tuning support, and Agent Skills.
🔗 GitHub: github.com/OpenBMB/MiniCPM-O…
🤗 Hugging Face:
MiniCPM-OCR 1.0：huggingface.co/openbmb/MiniC…
MiniCPM-OCR Jev：huggingface.co/openbmb/MiniC…
👉 Try the Demo：platform.modelbest.ai/consol…
Use the API：platform.modelbest.ai/consol…
Anderson retweeted
describe what you want built, not every step.
that’s the workflow i want to explore with coding agents.
qoder lets you hand off tasks and track the agent. deepseek-flash is free within the daily limit through oct 31.
qoder.com/qoder
🤝 Paid partnership
Another open source model worth keeping an eye on. 👀
Ling-3.1-flash-1019 scored 44 on the Artificial Analysis Intelligence Index, up from 41, while improving on several reasoning and agent benchmarks.
And now there's a base model too. That's a big opportunity for developers.
Anderson retweeted
my favourite genre of agent bug is the one that's perfectly polite, perfectly on brand and sent to the same person every six minutes
🤝 Paid partnership
Agents f*ck up.
When your SDR agent does, it shouldn't take the domain your invoices go out from with it.
OpenMail gives every agent its own inbox, its own sending limits and its own reputation.
openmail.sh
OCR’s real headache isn’t reading pages it’s catching mistakes after they hit your workflow.
MiniCPM-OCR pairs a fast 1B OCR model with a separate quality-checker, making document AI more trustworthy.
Your OCR says it’s done—but did it get everything right? 🤔 In high-stakes workflows, even a small error can be costly.
🚀 Meet the MiniCPM-OCR suite: an open-source solution to parse fast and guard the results.
Two new models:
🔹 MiniCPM-OCR 1.0 — SOTA OCR performance and speed at 1B scale. Compact, accurate, and fast.
🔹 MiniCPM-OCR Jev — A 2B decision model that helps assess parsing accuracy, with performance approaching that of Qwen3.5-397B-A17B on JevOCRBench.
⚡ What’s new?
Small model, SOTA results: MiniCPM-OCR 1.0 scores 93.68 on Real5-OmniDocBench and 54.7 on OCRBench v2.
A sharp goalkeeper 🛡️: MiniCPM-OCR Jev is model-agnostic, checking text, formulas, and tables from different OCR parsers. It achieves 93.6 on JevOCRBench, outperforming JEV-35B-A3B (90.1) and approaching Qwen3.5-397B-A17B (93.8).
Up to 128 pages per minute 🤯: With DSpark, MiniCPM-OCR runs at 1.49× the speed of PaddleOCR-VL-1.6 and 3.62× that of MinerU2.5 Pro. The SDK’s concurrency optimizations further boost throughput to as much as 128 pages per minute.
Deploy where you work: Day-0 support for Arm, Houmo AI, Intel, and NVIDIA, plus 10 more chip vendors through FlagOS.
Ready for your stack: vLLM, SGLang, llama.cpp, and Ollama, with quantized models, fine-tuning support, and Agent Skills.
🔗 GitHub: github.com/OpenBMB/MiniCPM-O…
🤗 Hugging Face:
MiniCPM-OCR 1.0：huggingface.co/openbmb/MiniC…
MiniCPM-OCR Jev：huggingface.co/openbmb/MiniC…
👉 Try the Demo：platform.modelbest.ai/consol…
Use the API：platform.modelbest.ai/consol…
T1 going for four in a row. Gen.G in as the No. 1 seed. HLE holding MSI.
And a robotics data company bought a box at the final so 35 engineers who are cracked at League can watch it with the Mecka AI team.
This is crazyyy 😭 Who's taking it?
🤝 Paid partnership · 🤖 Made with AI
The league of legends championship is in the usa this year.. and we booked the largest private box in the stadium
@MeckaAI cofounder @jasontheutopian and first engineer Anthony, both ranked challenger as top 50 players in the world
So we’re inviting 30 elite engineers, researchers and operators to join us in the ownership suite to watch the finals front and center
Anyone with world-class skill in a video game, not just in league, send us your rank or evidence of exceptional ability in what you do
The championship is nov 14 at the barclays center in new york
Apply by Nov 8. Link in comments
describe what you want built, not every step.
that’s the workflow i want to explore with coding agents.
qoder lets you hand off tasks and track the agent. deepseek-flash is free within the daily limit through oct 31.
qoder.com/qoder
🤝 Paid partnership
my favourite genre of agent bug is the one that's perfectly polite, perfectly on brand and sent to the same person every six minutes
🤝 Paid partnership
Agents f*ck up.
When your SDR agent does, it shouldn't take the domain your invoices go out from with it.
OpenMail gives every agent its own inbox, its own sending limits and its own reputation.
openmail.sh
Anderson retweeted
A new open-source model is taking healthcare AI beyond basic Q&A.
Ling-3.0-flash-Sante improves complex medical reasoning, clinical tasks, and tool-assisted research, scoring 90.47 on multi-agent BrowseComp.
With public weights, developers can now build and test their own apps.
Today, we’re officially open-sourcing Ling-3.0-flash-Sante.
Sante is a healthcare-enhanced MoE model with 124B total parameters and only 5.1B active parameters per token, delivering leading performance among similarly sized models on professional healthcare tasks and complex medical reasoning.
Beyond answering medical questions, Sante is built for real-world healthcare research and workflows, with strengthened capabilities in deep research, multi-agent collaboration, and tool use.
The model weights are now available — we welcome researchers, developers, and healthcare practitioners to explore Sante and build with it.
More below ↓
A new open-source model is taking healthcare AI beyond basic Q&A.
Ling-3.0-flash-Sante improves complex medical reasoning, clinical tasks, and tool-assisted research, scoring 90.47 on multi-agent BrowseComp.
With public weights, developers can now build and test their own apps.
Today, we’re officially open-sourcing Ling-3.0-flash-Sante.
Sante is a healthcare-enhanced MoE model with 124B total parameters and only 5.1B active parameters per token, delivering leading performance among similarly sized models on professional healthcare tasks and complex medical reasoning.
Beyond answering medical questions, Sante is built for real-world healthcare research and workflows, with strengthened capabilities in deep research, multi-agent collaboration, and tool use.
The model weights are now available — we welcome researchers, developers, and healthcare practitioners to explore Sante and build with it.
More below ↓
Anderson retweeted
Freelancing looks like freedom until you’re buried in proposals, follow-ups, deadlines, approvals, and invoices.
@_MonoDesk keeps your entire client workflow in one place, so projects move forward without app-hopping.
🤝 Paid partnership
You know us as your creative asset toolkit.
Now meet @_MonoDesk, a whole new app from our team, built to handle the boring bits of creative work.
Less time on proposals, chasing approvals and sending invoices. More time on the creative side of your business.
Anderson retweeted
Maxinsights films people working at normal speed in real homes and hotels.
Seems way more useful for a robot to watch that than some staged mock room.
🤝 Paid partnership
Robots learn by watching people work. Most of what they watch is staged.
At @Maxinsights, we capture real work at natural speed in real homes, restaurants, and hotels. 2M+ hours so far.
Today we’re announcing a $30M seed round led by @PeakXVPartners, plus two new products:
MaxLoop: same-day quality scores on every clip
MaxGlove: touch and force data a camera can't capture
Thank you to our investors, builders, operators and customers. More to come.
Anderson retweeted
Everyone talks about smarter AI. Too little about the data behind it.
Private codebases, real task environments, internal docs, specialized datasets data public pipelines can’t reach.
@DataVendor_ai is building a marketplace to connect that data with the AI labs that need it.
🤝 Paid partnership
We're announcing the public beta of datavendor.ai
We spent the last two years building the best solution for people buying and selling data, whether you're a researcher in a lab or a company trying to sell data.
Data is the fundamental bottleneck for AI progress, and DataVendor will be the infra that solves it.
Anderson retweeted
People talk about your brand every day—in comments, group chats, and DMs.
Do you see any of it?
@nectarsocialai uses AI agents to surface hidden mentions, find creators, spot trends, and help you engage driving $150M in attributed social revenue.
🤝 Paid partnership
Engineers stopped writing every line of code and started directing agents.
Growth and marketing teams are in that moment now.
Meet the next chapter of Nectar: the agentic growth stack. One platform replacing 5+ legacy tools. Agents that work every message, creator outreach and creative research about your brand, and tie it all back to revenue 🍊
Anderson retweeted
The AI race isn’t just bigger models it’s better results.
Ling-3.1-flash-1019 raises its Intelligence Index to 44 and boosts agentic benchmark scores.
Open weights are coming, giving developers more flexibility to build, test, and deploy custom AI workflows.
Freelancing looks like freedom until you’re buried in proposals, follow-ups, deadlines, approvals, and invoices.
@_MonoDesk keeps your entire client workflow in one place, so projects move forward without app-hopping.
🤝 Paid partnership
You know us as your creative asset toolkit.
Now meet @_MonoDesk, a whole new app from our team, built to handle the boring bits of creative work.
Less time on proposals, chasing approvals and sending invoices. More time on the creative side of your business.
Maxinsights films people working at normal speed in real homes and hotels.
Seems way more useful for a robot to watch that than some staged mock room.
🤝 Paid partnership
Robots learn by watching people work. Most of what they watch is staged.
At @Maxinsights, we capture real work at natural speed in real homes, restaurants, and hotels. 2M+ hours so far.
Today we’re announcing a $30M seed round led by @PeakXVPartners, plus two new products:
MaxLoop: same-day quality scores on every clip
MaxGlove: touch and force data a camera can't capture
Thank you to our investors, builders, operators and customers. More to come.
Everyone talks about smarter AI. Too little about the data behind it.
Private codebases, real task environments, internal docs, specialized datasets data public pipelines can’t reach.
@DataVendor_ai is building a marketplace to connect that data with the AI labs that need it.
🤝 Paid partnership
We're announcing the public beta of datavendor.ai
We spent the last two years building the best solution for people buying and selling data, whether you're a researcher in a lab or a company trying to sell data.
Data is the fundamental bottleneck for AI progress, and DataVendor will be the infra that solves it.
People talk about your brand every day—in comments, group chats, and DMs.
Do you see any of it?
@nectarsocialai uses AI agents to surface hidden mentions, find creators, spot trends, and help you engage driving $150M in attributed social revenue.
🤝 Paid partnership
Engineers stopped writing every line of code and started directing agents.
Growth and marketing teams are in that moment now.
Meet the next chapter of Nectar: the agentic growth stack. One platform replacing 5+ legacy tools. Agents that work every message, creator outreach and creative research about your brand, and tie it all back to revenue 🍊