Ulobex  @Ulobex

Full-Stack Engineer | AI Tools & Emerging Tech Analyst. Breaking down global trends in AI and tools. DM for Collabs passionfroot.me/ulobex USA Joined July 2020

Tweets

13K
Followers

6K
Following

2K
Likes

12K

Ulobex  @Ulobex

19 minutes ago

@Kanojiyaaakash1 Open models closing the gap so fast Verdent plus Kimi

0 0 0 3 0

View Details

Ulobex  @Ulobex

47 minutes ago

@KaiteeShiks The citations are what sold me. Much easier to verify everything.

0 0 6 161 0

View Details

Ulobex  @Ulobex

a day ago

@KaiteeShiks I like that it supports people instead of replacing them.

0 0 6 14 0

View Details

5/ If you're curious what agentic AI actually looks like in practice, Accio Work is worth exploring. You can download it below and try the workflow yourself. accio.com/work/app?defau… #AccioWork #MyAccioWorks #AgenticBusiness #AIAgent

0 0 8 6K 4

View Details

Ulobex  @Ulobex

a day ago

4/ What makes this different isn't that it can answer more questions. It's that it focuses on execution. The combination of Agents and Plugins means repetitive business tasks can actually move forward while you focus on higher-value work.

1 0 8 6K 5

View Details

Ulobex  @Ulobex

a day ago

Most AI tools stop after giving you an answer. Accio Work goes a step further. Instead of just generating ideas, it builds a team of autonomous agents that can source products, compare suppliers, prepare inquiries, and execute real business workflows. Here's what stood out while testing it.

14 132 146 45K 12

View Details

Ulobex  @Ulobex

a day ago

@GesoraMeshack @TopviewAIhq This feels much closer to filmmaking than prompting.

0 0 3 12 0

View Details

Ulobex  @Ulobex

2 days ago

@Iriss_Quinn Supporting more languages is great, but quality has to come first.

0 0 2 11 0

View Details

Ulobex  @Ulobex

2 days ago

@GesoraMeshack Personality is much harder to preserve than vocabulary.

0 0 1 18 0

View Details

Ulobex  @Ulobex

2 days ago

@KaiteeShiks This feels like the direction the industry needed to move toward.

0 0 4 13 0

View Details

Ulobex  @Ulobex

2 days ago

@KaiteeShiks @ThineAI Qualifying leads is half the battle.

0 0 5 23 0

View Details

Ulobex  @Ulobex

2 days ago

@KaiteeShiks The time savings will add up quickly with this.

0 0 3 11 0

View Details

Ulobex  @Ulobex

5 days ago

Also worth pointing out that this isn't just one leaderboard. On Sierra's μ-Bench, Modulate also ranks #1 for Mandarin transcription, ahead of Google Chirp-3 and ElevenLabs. Different benchmark, same trend: research.sierra.ai/mubench/

0 0 7 8K 5

View Details

Ulobex  @Ulobex

5 days ago

I mentioned @Modulate_ai a while back because they felt like one of those small teams quietly building something interesting. Looks like that bet is paying off. They're now ranked #1 out of 88 models on Hugging Face's Open ASR Leaderboard, a public benchmark, ahead of Google, OpenAI Whisper variants, and ElevenLabs. It's rare to see an independent team climb to the top of a leaderboard like this, especially in a space dominated by much bigger companies. Worth watching if you're following speech AI. You can try it yourself with 1,000 free credits: modulate.ai/api/speech-to-…