Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

Mistral OCR 4 just dropped with bounding boxes (their most-requested feature) so I plugged it into my form-filling test as the helper model. Qwen3.6 reasons, Mistral localizes. Result? Boxes detected, fields filled, mostly landing in the lines. Not pixel-perfect. But close? Yeah, I'll call it close.

stevibe's profile picture

stevibe

26,996 subscribers

40,997 Aufrufe • vor 18 Tagen •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

0 Kommentare

Keine Kommentare verfügbar

Kommentare vom Original-Post werden hier angezeigt

Ähnliche Videos

Mistral OCR 3 now returns confidence scores, at page level or word level. Know exactly how sure the model is about what it extracted. Available now via API and to test in Mistral Studio.
0:27

Mistral OCR 3 now returns confidence scores, at page level or word level. Know exactly how sure the model is about what it extracted. Available now via API and to test in Mistral Studio.

Mistral AI for Developers

29,944 Aufrufe • vor 3 Monaten

Introducing Mistral OCR 4:
3:13

Introducing Mistral OCR 4:

Mistral AI for Developers

162,214 Aufrufe • vor 18 Tagen

I designed a new test specifically for multimodal models: fill out a paper form. And it's much harder than it sounds. This isn't typing into an electronic field that captures your text. The form is just an image. The model has to place each form element: text, checkmarks — at the correct pixel position on the canvas itself. Results: 🟢 Kimi K2.6 → done in 3:45, 16.7k output tokens 🟡 Step 3.7 Flash → half the fields, 57k output tokens 🔴 Gemini 3.5 Flash → 489k output tokens, never finished. I had to kill it. Gemini burned ~29x more output tokens than Kimi on the exact same task, and Kimi's was the only form that actually looked filled out. The test, a mocked application form, contains some challenging parts, such as one-character-per-box fields. I provided every model the same set of tools: > get canvas size > drop probe markers to find coordinates > add text > add checkmarks > move elements > take a screenshot anytime to check their own work > ... etc So it's vision + spatial reasoning + tool use + long context, all at once. Small models (Qwen, Gemma) can't really complete this test, so I skipped them. What happened: > Kimi nailed name, DOB, ID, gender, marital status, nationality, email, phone, address, postal code — placement slightly loose, but content correct. 15 turns. Clean. > Step got maybe half right — fields dropped, "United States" landed in the email line, data floating outside boxes. Burned 1.24M input tokens doing it (81 turns of re-reading the canvas). > Gemini almost got there visually... then spiraled. By turn 40 it was issuing a delete_elements call wiping element IDs 365–425, basically erasing its own work. 31 minutes, 489k output tokens, still streaming. Terminated. The takeaway isn't "Gemini bad." This test is indeed difficult. But token efficiency is capability now. A model that needs 30x the tokens and still can't converge is going to be 30x the cost in production. Kimi K2.6 just quietly did the thing.
0:38

I designed a new test specifically for multimodal models: fill out a paper form. And it's much harder than it sounds. This isn't typing into an electronic field that captures your text. The form is just an image. The model has to place each form element: text, checkmarks — at the correct pixel position on the canvas itself. Results: 🟢 Kimi K2.6 → done in 3:45, 16.7k output tokens 🟡 Step 3.7 Flash → half the fields, 57k output tokens 🔴 Gemini 3.5 Flash → 489k output tokens, never finished. I had to kill it. Gemini burned ~29x more output tokens than Kimi on the exact same task, and Kimi's was the only form that actually looked filled out. The test, a mocked application form, contains some challenging parts, such as one-character-per-box fields. I provided every model the same set of tools: > get canvas size > drop probe markers to find coordinates > add text > add checkmarks > move elements > take a screenshot anytime to check their own work > ... etc So it's vision + spatial reasoning + tool use + long context, all at once. Small models (Qwen, Gemma) can't really complete this test, so I skipped them. What happened: > Kimi nailed name, DOB, ID, gender, marital status, nationality, email, phone, address, postal code — placement slightly loose, but content correct. 15 turns. Clean. > Step got maybe half right — fields dropped, "United States" landed in the email line, data floating outside boxes. Burned 1.24M input tokens doing it (81 turns of re-reading the canvas). > Gemini almost got there visually... then spiraled. By turn 40 it was issuing a delete_elements call wiping element IDs 365–425, basically erasing its own work. 31 minutes, 489k output tokens, still streaming. Terminated. The takeaway isn't "Gemini bad." This test is indeed difficult. But token efficiency is capability now. A model that needs 30x the tokens and still can't converge is going to be 30x the cost in production. Kimi K2.6 just quietly did the thing.

stevibe

25,304 Aufrufe • vor 1 Monat

I needed an app to export markdown from images and PDFs, so I built one using the best model on the market: Mistral OCR. It can convert images, PDFs, and even extract tables and images, keeping them contextually within the markdown. All you need is a Mistral AI API.
1:15

I needed an app to export markdown from images and PDFs, so I built one using the best model on the market: Mistral OCR. It can convert images, PDFs, and even extract tables and images, keeping them contextually within the markdown. All you need is a Mistral AI API.

Pietro Schirano

48,786 Aufrufe • vor 11 Monaten

Most people think a 0.8B model is a toy. 3rd test. This time: OCR I fed Qwen 3.5 0.8B three images: a Kindle page, my terrible handwriting x 2. Mostly correct. Running locally. From a model smaller than your browser tab.
0:33

Most people think a 0.8B model is a toy. 3rd test. This time: OCR I fed Qwen 3.5 0.8B three images: a Kindle page, my terrible handwriting x 2. Mostly correct. Running locally. From a model smaller than your browser tab.

stevibe

125,136 Aufrufe • vor 4 Monaten

Qwen3.6 35B-A3B dropped yesterday, so I ran it on 4 GPUs to see how it performs: 🟣 RTX 3090 — 49.78 tok/s, TTFT 852ms 🟡 RTX 4090 — 118.93 tok/s, TTFT 686ms 🟢 RTX 5090 — 160.37 tok/s, TTFT 409ms 🔵 DGX Spark — 59.98 tok/s, TTFT 228ms I went with ollama as the backend because honestly, it's the easiest way for most people to get started. One command, model pulled, done. I used Q4_K_M (24GB) across all four cards. The reason is the 3090 and 4090 don't support NVFP4 (only the 5090 and DGX Spark could use it). Keeping the same quant everywhere felt like the fairest way to compare. And yes, you can absolutely squeeze more performance out of every card with vLLM, SGLang, or TensorRT-LLM. But that's not what this test is about. This is just the out-of-the-box experience for folks who own a GPU and want to try the new model tonight.
1:24

Qwen3.6 35B-A3B dropped yesterday, so I ran it on 4 GPUs to see how it performs: 🟣 RTX 3090 — 49.78 tok/s, TTFT 852ms 🟡 RTX 4090 — 118.93 tok/s, TTFT 686ms 🟢 RTX 5090 — 160.37 tok/s, TTFT 409ms 🔵 DGX Spark — 59.98 tok/s, TTFT 228ms I went with ollama as the backend because honestly, it's the easiest way for most people to get started. One command, model pulled, done. I used Q4_K_M (24GB) across all four cards. The reason is the 3090 and 4090 don't support NVFP4 (only the 5090 and DGX Spark could use it). Keeping the same quant everywhere felt like the fairest way to compare. And yes, you can absolutely squeeze more performance out of every card with vLLM, SGLang, or TensorRT-LLM. But that's not what this test is about. This is just the out-of-the-box experience for folks who own a GPU and want to try the new model tonight.

stevibe

398,628 Aufrufe • vor 2 Monaten

Fable 5 just made the best Remotion video I have ever generated with ANY model. It is not close. It cloned BridgeSpace pixel perfect. It nailed every selling point. It designed the motion, the pacing, the whole thing. The design taste and creativity of this model are simply unmatched. Watch this and remember: an AI made this.
1:31

Fable 5 just made the best Remotion video I have ever generated with ANY model. It is not close. It cloned BridgeSpace pixel perfect. It nailed every selling point. It designed the motion, the pacing, the whole thing. The design taste and creativity of this model are simply unmatched. Watch this and remember: an AI made this.

BridgeMind

77,912 Aufrufe • vor 9 Tagen

The best way to bring the composition from your head into an image → Ideogram V4 + drawing bounding boxes in Comfy. The control here is quite unique. The model uses structured JSON so drawing bounding boxes to get the exact placement works very well. The model only needs 12 steps (turbo), so iterating with different seeds + very impressive text rendering capability leads me to say this is a state of the art open source image model right now. Using the 'Ideogram 4 Prompt Builder' node by Kijai.
0:49

The best way to bring the composition from your head into an image → Ideogram V4 + drawing bounding boxes in Comfy. The control here is quite unique. The model uses structured JSON so drawing bounding boxes to get the exact placement works very well. The model only needs 12 steps (turbo), so iterating with different seeds + very impressive text rendering capability leads me to say this is a state of the art open source image model right now. Using the 'Ideogram 4 Prompt Builder' node by Kijai.

rob - comfyui

24,794 Aufrufe • vor 1 Monat

What does it take to bring open models into the enterprise? On the NVIDIA AI Podcast, Mistral CTO and co-founder Timothée Lacroix joins to discuss Mistral AI’s open-model philosophy, its Forge customization framework, and its collaboration with NVIDIA through the Nemotron Coalition.
0:31

What does it take to bring open models into the enterprise? On the NVIDIA AI Podcast, Mistral CTO and co-founder Timothée Lacroix joins to discuss Mistral AI’s open-model philosophy, its Forge customization framework, and its collaboration with NVIDIA through the Nemotron Coalition.

NVIDIA

27,138 Aufrufe • vor 1 Monat

The #SPX6900 thesis is playing out in real time: “The perfect meme token.” “It just checks off so many boxes.” /////////////////////////////////////////////////////// Stop trading and believe in something. @SPX6900
1:26

The #SPX6900 thesis is playing out in real time: “The perfect meme token.” “It just checks off so many boxes.” /////////////////////////////////////////////////////// Stop trading and believe in something. @SPX6900

0xpepethefrog 💹🧲

100,086 Aufrufe • vor 1 Jahr

🚨| Sir Lewis Hamilton's post race interview. "Good result, amazing result for the team. Third and fourth, good points." "I went forwards even after a bad start, so lots and lots of positives to take from it. I think there's, yeah, naturally I want to be further up, so just got to keep working at it." "Considering we've not been upgrading the car for us to be kind of in the fight, it shows that the team's done a really, really great job this weekend. I hope everyone back at the factory is happy. And yeah, we're definitely on the right path. I'm so close, still not with that podium, but I'll keep trying." - F1 2025 USGP -
1:39

🚨| Sir Lewis Hamilton's post race interview. "Good result, amazing result for the team. Third and fourth, good points." "I went forwards even after a bad start, so lots and lots of positives to take from it. I think there's, yeah, naturally I want to be further up, so just got to keep working at it." "Considering we've not been upgrading the car for us to be kind of in the fight, it shows that the team's done a really, really great job this weekend. I hope everyone back at the factory is happy. And yeah, we're definitely on the right path. I'm so close, still not with that podium, but I'll keep trying." - F1 2025 USGP -

sim

23,376 Aufrufe • vor 8 Monaten

Frankston is one of the most searched suburbs for homebuyers. And there's little surprise why. It ticks a lot of boxes - close to public transport and services. It's the perfect place for one of our Activity Centres - where we'll build help more homes, close to where people want to live.
0:45

Frankston is one of the most searched suburbs for homebuyers. And there's little surprise why. It ticks a lot of boxes - close to public transport and services. It's the perfect place for one of our Activity Centres - where we'll build help more homes, close to where people want to live.

Jacinta Allan

35,305 Aufrufe • vor 1 Jahr

bored as shit so i gave it a shot. not really perfect but yeah whatever
2:12

bored as shit so i gave it a shot. not really perfect but yeah whatever

CamaCon

18,173 Aufrufe • vor 11 Monaten

We used Mistral OCR with Claude 3.7 to create blog-style overviews for arXiv papers Generate beautiful research blogs with figures, key insights, and clear explanations from the paper with just one click Understand papers in minutes - not hours
0:25

We used Mistral OCR with Claude 3.7 to create blog-style overviews for arXiv papers Generate beautiful research blogs with figures, key insights, and clear explanations from the paper with just one click Understand papers in minutes - not hours

alphaXiv

153,686 Aufrufe • vor 1 Jahr

Experiment with Voxtral TTS directly in the Mistral Studio playground. Select one of the Mistral voices or record your own.
2:36

Experiment with Voxtral TTS directly in the Mistral Studio playground. Select one of the Mistral voices or record your own.

Mistral AI

56,849 Aufrufe • vor 3 Monaten

Baier: Why not just hand them over? Trump: Because I had boxes, I want to go through the boxes and get all my personal things
1:00

Baier: Why not just hand them over? Trump: Because I had boxes, I want to go through the boxes and get all my personal things

Acyn

2,785,167 Aufrufe • vor 3 Jahren