Video yükleniyor...

Video Yüklenemedi

Ana Sayfaya Dön

OpenAI Codex-Spark powered by Cerebras You can now just build things faster—at 1,000 tokens/s.

cerebras's profile picture

Cerebras

76,105 subscribers

287,907 görüntüleme • 7 ay önce •via X (Twitter)

Bilim & Teknoloji
LIVE

Anya Rossi• Live Now

Private livecam show

32 Yorum

Carlo Taleon profil fotoğrafı
Carlo Taleon7 ay önce

Is 5.3 Codex Spark a "faster but dumber" or "just faster" version of 5.3 Codex?

Joe Williams profil fotoğrafı
Joe Williams7 ay önce

We want 4o! Give us 4o! #keep4o

nick profil fotoğrafı
nick7 ay önce

1000 tokens/s, nice. now we just need to worry about ai writing code that's *beautifully* buggy at scale.

EL profil fotoğrafı
EL7 ay önce

Finally, my 'it's inferencing' excuse is obsolete. Thanks Cerebras.

Bill profil fotoğrafı
Bill7 ay önce

Will you guys be releasing anymore code plans / is there any way to get coupons to get in still. I’ve been trying for months but always sold out

HKay profil fotoğrafı
HKay7 ay önce

I thought we had to wait until 2028 to get this much power?

Dilreet Raju profil fotoğrafı
Dilreet Raju7 ay önce

Most likely codex spark is a distilled version or quantized version of codex with 5.3. Smaller weights to fit on the cerebras SRAM. Hopefully not too many performance issues with the model, but 1000 tps might make it worth it. OpenAI is def the best AI lab at distillations so I think it has potential to be a solid product. Good job getting it out there and moving fast @cerebras

Isaac Kargar profil fotoğrafı
Isaac Kargar7 ay önce

We need 5.3 codex xhigh witg this speed ...

Devesh Yadav profil fotoğrafı
Devesh Yadav7 ay önce

1000 tokens/s is crazy

Val profil fotoğrafı
Val7 ay önce

hi pls add glm 5

Said profil fotoğrafı
Said7 ay önce

crazy fast omg!!

Hamza profil fotoğrafı
Hamza7 ay önce

congrats guy for this

Harry Stuckler profil fotoğrafı
Harry Stuckler7 ay önce

Awesome Work - Congratulations everyone.

Clawglasses 🪽 profil fotoğrafı
Clawglasses 🪽7 ay önce

finally feels like the ai is keeping up instead of making me wait

luis profil fotoğrafı
luis7 ay önce

Lol @AndrewCurran_

Marrick profil fotoğrafı
Marrick7 ay önce

Yay super fast sloppy garbage

YEETUS DELETUS profil fotoğrafı
YEETUS DELETUS7 ay önce

Where three coupons drops at for Cerebras Code Max 😭😭😭😭😭😭

Juan profil fotoğrafı
Juan7 ay önce

no fucking way it was this simple

Neil profil fotoğrafı
Neil7 ay önce

whats was t/s before? @grok

Nico profil fotoğrafı
Nico7 ay önce

dam yall have been cooking🔥

Hypocrisy profil fotoğrafı
Hypocrisy7 ay önce

useless if there is no api so the public can use it. Openai has been extremely frugal in releasing open api's lately. The competition will catch up in less than 2 months as always and it will be open source.

Tributary profil fotoğrafı
Tributary7 ay önce

Amazing, awesome to see this partnership!

Quantum Flux profil fotoğrafı
Quantum Flux7 ay önce

So um care to tell us what it is?

James profil fotoğrafı
James7 ay önce

best of n?

Pavel Moore profil fotoğrafı
Pavel Moore7 ay önce

awesome and thank you

wenlong profil fotoğrafı
wenlong7 ay önce

Token reburning🔥🔥🔥

NewAIWorld profil fotoğrafı
NewAIWorld7 ay önce

Finally I can ship bugs faster than my brain can process regret

neamtu profil fotoğrafı
neamtu7 ay önce

wild!

Eddie profil fotoğrafı
Eddie7 ay önce

so they say 1k tokens/s is revolutionary, but who’s counting when the whole thing feels like chasing shadows in a cave

KING 👑 profil fotoğrafı
KING 👑7 ay önce

Contract address ?

sonald profil fotoğrafı
sonald7 ay önce

So the new bottleneck is how fast we can think. Vibecoding just turned into speedcoding. when will api available ?

-__- profil fotoğrafı
-__-7 ay önce

Where you at @firstadopter

Benzer Videolar

Our coding workflows were designed to accommodate slow inference. OpenAI's Codex Spark powered by Cerebras changes the game. Here's how we make the most out of 1,200 tokens per second, with Sarah Chieng.
5:26

Our coding workflows were designed to accommodate slow inference. OpenAI's Codex Spark powered by Cerebras changes the game. Here's how we make the most out of 1,200 tokens per second, with Sarah Chieng.

Cerebras

13,421 görüntüleme • 7 ay önce

🚨 Cerebras Inference is now 3x faster: Llama3.1-70B just broke 2,100 tokens/s - 16x faster than the fastest GPU solution - 8x faster than GPUs running Llama *3B* - It's like the perf of a new hardware generation in a single software release Available now at
0:26

🚨 Cerebras Inference is now 3x faster: Llama3.1-70B just broke 2,100 tokens/s - 16x faster than the fastest GPU solution - 8x faster than GPUs running Llama *3B* - It's like the perf of a new hardware generation in a single software release Available now at

Cerebras

236,476 görüntüleme • 1 yıl önce

Introducing Cerebras Inference ‣ Llama3.1-70B at 450 tokens/s – 20x faster than GPUs ‣ 60c per M tokens – a fifth the price of hyperscalers ‣ Full 16-bit precision for full model accuracy ‣ Generous rate limits for devs Try now:
0:21

Introducing Cerebras Inference ‣ Llama3.1-70B at 450 tokens/s – 20x faster than GPUs ‣ 60c per M tokens – a fifth the price of hyperscalers ‣ Full 16-bit precision for full model accuracy ‣ Generous rate limits for devs Try now:

Cerebras

708,154 görüntüleme • 2 yıl önce

We built Money Agent, a personal finance assistant powered by Qwen 3.8 27B on Cerebras. Now you can turn a home-buying question into a real-time conversation, then into a financial goal, faster.
0:49

We built Money Agent, a personal finance assistant powered by Qwen 3.8 27B on Cerebras. Now you can turn a home-buying question into a real-time conversation, then into a financial goal, faster.

Cerebras

107,457 görüntüleme • 20 gün önce

what can you build with gpt-5.3-codex-spark? jason liu from OpenAI demos 3 real workflows — ones you can set up yourself inside the Codex app to help you spend less time on overhead and more time building. 00:09 – what is gpt-5.3-codex-spark? 00:25 – workflow 1: multi-agent daily briefing from slack, drive & meets 01:06 – workflow 2: automated PR review 01:31 – workflow 3: real-time interactive coding 02:56 – what speed changes, and what's coming next
3:45

what can you build with gpt-5.3-codex-spark? jason liu from OpenAI demos 3 real workflows — ones you can set up yourself inside the Codex app to help you spend less time on overhead and more time building. 00:09 – what is gpt-5.3-codex-spark? 00:25 – workflow 1: multi-agent daily briefing from slack, drive & meets 01:06 – workflow 2: automated PR review 01:31 – workflow 3: real-time interactive coding 02:56 – what speed changes, and what's coming next

Cerebras

23,633 görüntüleme • 6 ay önce

Introducing ⚡️Cerebras MCP Server ⚡️ You can now turbocharge any AI editor that uses MCP for tool calling with 20x faster inference. Here is Claude Code writing files at breakneck speed via Cerebras. Get started now 👇
0:34

Introducing ⚡️Cerebras MCP Server ⚡️ You can now turbocharge any AI editor that uses MCP for tool calling with 20x faster inference. Here is Claude Code writing files at breakneck speed via Cerebras. Get started now 👇

Cerebras

42,338 görüntüleme • 1 yıl önce

GLM-4.6 is now live on Cerebras 🟧 - 1,000 TPS: 17x faster than Claude - World’s #1 tool calling model - Tasteful and reliable codegen - Self-serve API starting at $10 Try it now:
0:36

GLM-4.6 is now live on Cerebras 🟧 - 1,000 TPS: 17x faster than Claude - World’s #1 tool calling model - Tasteful and reliable codegen - Self-serve API starting at $10 Try it now:

Cerebras

33,135 görüntüleme • 10 ay önce

Codex can now help you build AI apps and agents faster with OpenAI APIs using the OpenAI Developers plugin.
0:40

Codex can now help you build AI apps and agents faster with OpenAI APIs using the OpenAI Developers plugin.

OpenAI Developers

320,090 görüntüleme • 5 ay önce

Cerebras Code: 20x faster than Claude, 1x the price Today we are launching two monthly coding plans: ➡️Cerebras Code Pro: $50/m – for indie developers ➡️Cerebras Code Max: $200/m – for power users with 5x rate limits Both plans get: Qwen3-Coder at 2,000 tokens/s, 131K context, and no weekly limits. Sign up now:
0:43

Cerebras Code: 20x faster than Claude, 1x the price Today we are launching two monthly coding plans: ➡️Cerebras Code Pro: $50/m – for indie developers ➡️Cerebras Code Max: $200/m – for power users with 5x rate limits Both plans get: Qwen3-Coder at 2,000 tokens/s, 131K context, and no weekly limits. Sign up now:

Cerebras

461,574 görüntüleme • 1 yıl önce

Can AI really replace SaaS? We asked Codex Spark to build a simple Salesforce clone. It generated a build plan, then wrote all the code in 29 seconds. And it works! Add contacts, live search, full pipeline dashboard – all unit tests passed. The same task on full Codex took ~5× longer. For rapid prototyping and self-contained work, nothing beats Spark. ✨
1:45

Can AI really replace SaaS? We asked Codex Spark to build a simple Salesforce clone. It generated a build plan, then wrote all the code in 29 seconds. And it works! Add contacts, live search, full pipeline dashboard – all unit tests passed. The same task on full Codex took ~5× longer. For rapid prototyping and self-contained work, nothing beats Spark. ✨

Cerebras

16,831 görüntüleme • 6 ay önce

Previewing Ultrafast mode for OpenAI's GPT 5.6 Sol, powered by Cerebras. GPT-5.6 Sol Ultrafast generates responses at up to 750 tokens per second. That's the full, GPT-5.6-Sol model -- up to 14× faster than the same model on Standard processing. It speedran Humanity’s Last Exam in 11h 11m, nearly 7× faster than Claude Fable 5 with comparable accuracy.
0:46

Previewing Ultrafast mode for OpenAI's GPT 5.6 Sol, powered by Cerebras. GPT-5.6 Sol Ultrafast generates responses at up to 750 tokens per second. That's the full, GPT-5.6-Sol model -- up to 14× faster than the same model on Standard processing. It speedran Humanity’s Last Exam in 11h 11m, nearly 7× faster than Claude Fable 5 with comparable accuracy.

Cerebras

1,291,350 görüntüleme • 1 ay önce

Ultrafast mode for GPT-5.6 Sol is now in limited preview, powered by Cerebras. We gave OpenAI's GPT-5.6 Sol the same prompt on Ultrafast and Standard: build a financial terminal-style dashboard for analysts. Ultrafast: 1 min 50 seconds Standard: 12 min 20 seconds Same result, nearly 7x faster.
0:51

Ultrafast mode for GPT-5.6 Sol is now in limited preview, powered by Cerebras. We gave OpenAI's GPT-5.6 Sol the same prompt on Ultrafast and Standard: build a financial terminal-style dashboard for analysts. Ultrafast: 1 min 50 seconds Standard: 12 min 20 seconds Same result, nearly 7x faster.

Cerebras

145,979 görüntüleme • 1 ay önce

"I used a billion tokens this week. I'm not even in the top 100 Codex users at OpenAI." We sat down with jason (creator of Instructor, now on OpenAI's Developer Experience team) to talk about how zero-latency inference is changing the way engineers work.
0:37

"I used a billion tokens this week. I'm not even in the top 100 Codex users at OpenAI." We sat down with jason (creator of Instructor, now on OpenAI's Developer Experience team) to talk about how zero-latency inference is changing the way engineers work.

Cerebras

93,023 görüntüleme • 5 ay önce

Introducing SuperNinja – the world’s fastest deep research by NinjaTech AI powered by Cerebras: ‣ 1-2min for full answer instead of 10min ‣ 5x faster than o3, Gemini Pro etc. ‣ Same accuracy as measured by GAIA
0:47

Introducing SuperNinja – the world’s fastest deep research by NinjaTech AI powered by Cerebras: ‣ 1-2min for full answer instead of 10min ‣ 5x faster than o3, Gemini Pro etc. ‣ Same accuracy as measured by GAIA

Cerebras

25,541 görüntüleme • 1 yıl önce

Gemma 4 31B is now available in Public Preview on Cerebras. Our first multimodal model runs at over 1,800 tokens/s for ultra-fast image and text workflows. Give it a try:
2:08

Gemma 4 31B is now available in Public Preview on Cerebras. Our first multimodal model runs at over 1,800 tokens/s for ultra-fast image and text workflows. Give it a try:

Cerebras

282,043 görüntüleme • 3 ay önce

We're excited to collaborate with Upstage, one of Korea's leading AI companies. Upstage's Solar 31B is now running on Cerebras at up to 2,000 tokens/sec. Watch Solar 31B complete a deep research query across 246 sources...how much faster do you think it is?
0:54

We're excited to collaborate with Upstage, one of Korea's leading AI companies. Upstage's Solar 31B is now running on Cerebras at up to 2,000 tokens/sec. Watch Solar 31B complete a deep research query across 246 sources...how much faster do you think it is?

Cerebras

39,293 görüntüleme • 3 ay önce

I gave both Codex and Claude Code full access to my twitter account. Codex spark (powered by Cerebras) did a full GTM outreach workflow 3x FASTER than Claude code, and used way less tool calls.
3:39

I gave both Codex and Claude Code full access to my twitter account. Codex spark (powered by Cerebras) did a full GTM outreach workflow 3x FASTER than Claude code, and used way less tool calls.

Sarah Chieng

75,746 görüntüleme • 7 ay önce

You can just build things.
1:00

You can just build things.

OpenAI

3,024,230 görüntüleme • 8 ay önce

The Fastest AI Just Got Faster. Meet CS-4.
1:08

The Fastest AI Just Got Faster. Meet CS-4.

Cerebras

1,538,687 görüntüleme • 1 ay önce

“We coined this term called the semi-async valley of death.” At Cerebras Supernova, Silas Alberti unpacked why coding tasks that take 1 to 3 minutes break flow, and how Cerebras powers Cognition’s SWE-1.7 in Devin at up to 1,000 tokens per second to keep developers in the loop.
1:27

“We coined this term called the semi-async valley of death.” At Cerebras Supernova, Silas Alberti unpacked why coding tasks that take 1 to 3 minutes break flow, and how Cerebras powers Cognition’s SWE-1.7 in Devin at up to 1,000 tokens per second to keep developers in the loop.

Cerebras

89,992 görüntüleme • 1 ay önce