Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

✅ clef-flash: fast + just $0.09/M input tokens ✅ clef: 27B decision model at $0.24/M input tokens ✅ clef-omni: image, audio + video inputs !!!!! All available on Cloudflare Workers AI! Let me show you all 3 using Pictionary.. Or to hear me make animal sounds...?

Jilles's profile picture

Jilles Soeters

4,253 subscribers

10,666 Aufrufe • vor 1 Monat •via X (Twitter)

Wissenschaft & TechnologieBildung
LIVE

Anya Rossi• Live Now

Private livecam show

5 Kommentare

Profilbild von Fayaz Ahmed
Fayaz Ahmedvor 1 Monat

Thats a circle?

Profilbild von Jilles Soeters
Jilles Soetersvor 1 Monat

Read more Changelog Hugging Face

Profilbild von Jilles Soeters
Jilles Soetersvor 1 Monat

CORRECTION: clef-flash is $0.038 per M input tokens. So 2.37x cheaper than I said ….

Profilbild von Tom Mann
Tom Mannvor 1 Monat

That's cool! Haven't tried Jev much but since I already use cloudflare might as well

Profilbild von B. Elliott Baker
B. Elliott Bakervor 1 Monat

this is insane.

Ähnliche Videos

Drex Max is here: our multimodal decision model text, images, and video 34 ms/decision, under 1s at 262K tokens $0.06 / 1M input tokens available today at
0:30

Drex Max is here: our multimodal decision model text, images, and video 34 ms/decision, under 1s at 262K tokens $0.06 / 1M input tokens available today at

Nace AI

232,127 Aufrufe • vor 26 Tagen

in honor of 20 year 30 rock anniversary + clef-omni release, decided to build a little experiment with workers + @cloudflare realtimekit "jenna's side... i'm not hearing it liz" try it yourself 👉
0:36

in honor of 20 year 30 rock anniversary + clef-omni release, decided to build a little experiment with workers + @cloudflare realtimekit "jenna's side... i'm not hearing it liz" try it yourself 👉

rita kozlov 🐀

12,213 Aufrufe • vor 1 Monat

We made it possible to train your own Decision model locally on just 3GB VRAM! We converted Qwen, Gemma, Llama all into decision models by fine-tuning using a Clef head, boosting accuracy from 30 to up to 78%. You can try it yourself via Unsloth Desktop!
0:31

We made it possible to train your own Decision model locally on just 3GB VRAM! We converted Qwen, Gemma, Llama all into decision models by fine-tuning using a Clef head, boosting accuracy from 30 to up to 78%. You can try it yourself via Unsloth Desktop!

Daniel Han

27,256 Aufrufe • vor 1 Monat

Realtime AI Conversations are here! Introducing PlayHT 2.0 Turbo ⚡️ Our new blazing fast Conversational AI Text-to-Speech model with <300ms latency! ✅ Input text streaming from LLMs ✅ Output audio streaming ✅ Clone any voice & accent Try here -
0:40

Realtime AI Conversations are here! Introducing PlayHT 2.0 Turbo ⚡️ Our new blazing fast Conversational AI Text-to-Speech model with <300ms latency! ✅ Input text streaming from LLMs ✅ Output audio streaming ✅ Clone any voice & accent Try here -

PlayAI

392,583 Aufrufe • vor 3 Jahren

OpenAI Decisions API と Clef-Flash を対決させてみた。Clef でリベンジさせた方がいいかな クイズ画像は #AIニケちゃん のふたり
2:18

OpenAI Decisions API と Clef-Flash を対決させてみた。Clef でリベンジさせた方がいいかな クイズ画像は #AIニケちゃん のふたり

あいびぃ

44,757 Aufrufe • vor 1 Monat

Theoretical counterplay and nerfs on Dragunov's G-Clef.
0:47

Theoretical counterplay and nerfs on Dragunov's G-Clef.

/𝗢𝗩𝗥𝗙𝗟𝗪

32,245 Aufrufe • vor 3 Monaten

De Clef sa wap pale? Clef ki bay Shakira hit, ki bay Beyonce hit, ki bay Avicii hit la? Wendyyy pa menm ka bay Thelo hit epi nou panse l ka superstar ke Clef? Nou pa respekte lejann menm non
0:26

De Clef sa wap pale? Clef ki bay Shakira hit, ki bay Beyonce hit, ki bay Avicii hit la? Wendyyy pa menm ka bay Thelo hit epi nou panse l ka superstar ke Clef? Nou pa respekte lejann menm non

Jeff Le Parolier (ADDM)

35,967 Aufrufe • vor 1 Jahr

Meet our new and fast 3D sculpting model! ✅ Single image to mesh with detailed geometry in the order of minutes (more GPUs coming to get to <30-60s) ✅ Dense mesh re-topology using our custom language model (minutes) Available on all Cube plans:
0:51

Meet our new and fast 3D sculpting model! ✅ Single image to mesh with detailed geometry in the order of minutes (more GPUs coming to get to <30-60s) ✅ Dense mesh re-topology using our custom language model (minutes) Available on all Cube plans:

Common Sense Machines

53,311 Aufrufe • vor 2 Jahren

Just listen to how crispy this game sounds with my audio. ✅ Make Your Gunshots Quiet ✅ Hear Footsteps Louder ✅ Improved Directionality in 7.1 ✅ Game Sounds like Call of Duty ❌ Sounds Like Tin Can
1:35

Just listen to how crispy this game sounds with my audio. ✅ Make Your Gunshots Quiet ✅ Hear Footsteps Louder ✅ Improved Directionality in 7.1 ✅ Game Sounds like Call of Duty ❌ Sounds Like Tin Can

Rara

81,923 Aufrufe • vor 5 Monaten

DeepSeek V4.1 Flash 1017 MAX vs Openai Gpt 6.1 Sol Extra High DeepSeek Input Tokens: 1.970.140 Output Tokens: 243.599 Time: 8 hours and 6 minutes Cost: $0.96 Gpt Sol Input Tokens: 96.063 Output Tokens: 54.730 Time: 20 minutes and 37 seconds Cost: $1,29 Which one do you think is the best?
0:30

DeepSeek V4.1 Flash 1017 MAX vs Openai Gpt 6.1 Sol Extra High DeepSeek Input Tokens: 1.970.140 Output Tokens: 243.599 Time: 8 hours and 6 minutes Cost: $0.96 Gpt Sol Input Tokens: 96.063 Output Tokens: 54.730 Time: 20 minutes and 37 seconds Cost: $1,29 Which one do you think is the best?

Rubens Soto | AI & SaaS

31,254 Aufrufe • vor 23 Tagen

Practice Drawing a Treble Clef #musictheory #musictheorytiktok #musictheoryforbeginners
0:26

Practice Drawing a Treble Clef #musictheory #musictheorytiktok #musictheoryforbeginners

Music Theory For Beginners

63,800 Aufrufe • vor 3 Jahren

Moondream 3.1 is available on Cloudflare Developers Workers AI Look how fricking fast this model is
0:32

Moondream 3.1 is available on Cloudflare Developers Workers AI Look how fricking fast this model is

Fayaz Ahmed

24,050 Aufrufe • vor 4 Monaten

Introducing SubQ Snap, the first long-context decision model for real enterprise work. Snap is the only decision model or LLM where compute per token stays constant as context grows. • 384K tokens of multimodal context, 12x larger than Jev. • 61.7 on Decision Index v0.2.1 vs Jev’s 57.9, outperforming on 13 of 16 enterprise benchmarks. •Processes 100K tokens and returns a decision in under one second. •$0.05/MM input tokens. Output is free. Available today to anyone. The first 10,000 sign-ups get 100M free input tokens. We’ll likely hit capacity fast and will add users as we add compute. Compute offers welcome 🙂. Latency will be variable as we tune the platform to this new model and category of users. Model speeds are 80ms for minimal context, 250ms for 32k, 660ms for 100k, and 2.5 seconds for 384k. Visit:
1:16

Introducing SubQ Snap, the first long-context decision model for real enterprise work. Snap is the only decision model or LLM where compute per token stays constant as context grows. • 384K tokens of multimodal context, 12x larger than Jev. • 61.7 on Decision Index v0.2.1 vs Jev’s 57.9, outperforming on 13 of 16 enterprise benchmarks. •Processes 100K tokens and returns a decision in under one second. •$0.05/MM input tokens. Output is free. Available today to anyone. The first 10,000 sign-ups get 100M free input tokens. We’ll likely hit capacity fast and will add users as we add compute. Compute offers welcome 🙂. Latency will be variable as we tune the platform to this new model and category of users. Model speeds are 80ms for minimal context, 250ms for 32k, 660ms for 100k, and 2.5 seconds for 384k. Visit:

Alexander Whedon

4,519,315 Aufrufe • vor 1 Monat

excited to announce my latest app Based Terminal: Explore and trade Base App coins all in one place ✨ ✅ fast native swaps across all tokens ✅ filter across market cap, high volume, holders, age, last trades & more coming Try it down below and let me know your thoughts👇
1:29

excited to announce my latest app Based Terminal: Explore and trade Base App coins all in one place ✨ ✅ fast native swaps across all tokens ✅ filter across market cap, high volume, holders, age, last trades & more coming Try it down below and let me know your thoughts👇

saint levan

46,494 Aufrufe • vor 1 Jahr

Google says its new AI model Gemini Omni can "create anything with any input." The first model in the Omni family, Gemini Omni Flash, launches today in the Gemini app, Google Flow, and on YouTube Shorts. #google #googleio
2:39

Google says its new AI model Gemini Omni can "create anything with any input." The first model in the Omni family, Gemini Omni Flash, launches today in the Gemini app, Google Flow, and on YouTube Shorts. #google #googleio

The Verge

240,800 Aufrufe • vor 5 Monaten

20.3 MILLION FREE TOKENS ON CLAUDE, GPT AND 30+ MODELS. 3 SITES. ONE OF THEM ONLY OPENS AT CERTAIN HOURS. > 5M tokens every month, 30+ models including the whole Claude lineup. Sign up, get all of them at once. > 15M tokens the moment you register. All current top models. Best input-token ratio of the three. > 300k tokens. Free Claude works only during certain hours. If the model is offline, just try later. Small providers, so don't put anything sensitive through them. Save this.
0:40

20.3 MILLION FREE TOKENS ON CLAUDE, GPT AND 30+ MODELS. 3 SITES. ONE OF THEM ONLY OPENS AT CERTAIN HOURS. > 5M tokens every month, 30+ models including the whole Claude lineup. Sign up, get all of them at once. > 15M tokens the moment you register. All current top models. Best input-token ratio of the three. > 300k tokens. Free Claude works only during certain hours. If the model is offline, just try later. Small providers, so don't put anything sensitive through them. Save this.

Atenov int.

104,380 Aufrufe • vor 1 Monat

Today we introduced a novel system one model, Clairic, based on regression instead of classification (i.e. decision models like Clef). This enables estimation of continuous values without having to exhaustively list choices upfront. It also enables very efficient extraction of structured data from unstructured inputs. Open weights coming soon!
1:22

Today we introduced a novel system one model, Clairic, based on regression instead of classification (i.e. decision models like Clef). This enables estimation of continuous values without having to exhaustively list choices upfront. It also enables very efficient extraction of structured data from unstructured inputs. Open weights coming soon!

Kevin Flansburg

129,483 Aufrufe • vor 26 Tagen

Meet Reka Edge – Our next-generation vision language model for physical AI. Uses 3x fewer input tokens and achieves 65% faster throughput compared to leading 8B models. Image understanding, video analysis, object detection, and tool use. Built for Action. Fast enough for production, deployable anywhere. Read more:
0:24

Meet Reka Edge – Our next-generation vision language model for physical AI. Uses 3x fewer input tokens and achieves 65% faster throughput compared to leading 8B models. Image understanding, video analysis, object detection, and tool use. Built for Action. Fast enough for production, deployable anywhere. Read more:

Reka

54,167 Aufrufe • vor 8 Monaten

La jeunesse est la clef pour faire partir Emmanuel Macron. Bloquez tout.
0:54

La jeunesse est la clef pour faire partir Emmanuel Macron. Bloquez tout.

Louis Boyard

31,669 Aufrufe • vor 1 Jahr

Ovi is out on Hugging Face Twin Backbone Cross-Modal Fusion for Audio-Video Generation Ovi is a veo-3 like, video+audio generation model that simultaneously generates both video and audio content from text or text+image inputs. Video+Audio Generation: Generate synchronized video and audio content simultaneously Flexible Input: Supports text-only or text+image conditioning 5-second Videos: Generates 5-second videos at 24 FPS, area of 720×720, at various aspect ratios (9:16, 16:9, 1:1, etc)
2:08

Ovi is out on Hugging Face Twin Backbone Cross-Modal Fusion for Audio-Video Generation Ovi is a veo-3 like, video+audio generation model that simultaneously generates both video and audio content from text or text+image inputs. Video+Audio Generation: Generate synchronized video and audio content simultaneously Flexible Input: Supports text-only or text+image conditioning 5-second Videos: Generates 5-second videos at 24 FPS, area of 720×720, at various aspect ratios (9:16, 16:9, 1:1, etc)

AK

23,082 Aufrufe • vor 1 Jahr