Manvinder @manv

San Francisco, CA Joined December 2008

Tweets

363
Followers

262
Following

308
Likes

204

Philipp Schmid @_philschmid

a year ago

Soft capping is now supported in Flash Attention! 🚀 Brace yourself for fast and accurate @GoogleDeepMind Gemma2 generations. 🏎️💥💨 Thank you @narsilou and @tri_dao ❤️

8 21 112 23K 26

Download Image

Gus (🤖🧠+🐍+🥑🗣️) @gusthema

a year ago

Hey everyone! 👋 Guess what? 💎Gemma 2 (9B)💎 is now available on Google AI Studio! Go play around with it and let me know what you think! 😄 aistudio.google.com/app/prompts/ne… #gemmaverse

1 4 21 2K 1

Danish Khan @nahkhsinad

a year ago

@MervinPraison @OpenAI @groq @GoogleAI @AnthropicAI @MistralAI @ollama @cohere @firecrawl_dev @AgentOpsAI Just went through the docs, this is best thing out there to experiment with AI agents on a low code mode. Executing a real office team/work scenario using ParisonAI!

0 1 1 395 0

Woosuk Kwon @woosuk_k

a year ago

Gemma 2 is also available in vLLM! 🎉github.com/vllm-project/v… Check out the update in the main branch and stay tuned for the next release coming soon

Google DeepMind @GoogleDeepMind

a year ago

Gemma 2 is also available in vLLM! 🎉github.com/vllm-project/v… Check out the update in the main branch and stay tuned for the next release coming soon

30 217 1K 469K 171

Download Video

3 23 139 15K 33

m_ric @AymericRoucher

a year ago

Google just released Gemma-2. The 27B version: ✅ directly becomes the best open-source LLM as per Chatbot Arena 🥊 Punches wayyy above its weight: I plotted Arena ELO vs model size below, it's crazy 🤯

2 5 36 12K 13

Download Image

Xavier(Xavi) Amatriain @xamat

a year ago

Gemma 2 27B beats Llama-3 70 on lmsys 🙌

Clément Farabet @clmt

a year ago

Gemma 2 27B beats Llama-3 70 on lmsys 🙌

11 79 325 113K 45

Download Image

0 1 5 733 0

The new Gemma-2 from @GoogleAI is optimized with @nvidia TensorRT-LLM or as an NVIDIA NIM inference microservice, with optimization for NVIDIA’s NeMo to come. LLM Inference can be up Up to 8x Faster on RTX With @nvidia's TensorRT-LLM 🔥 Main reason is unlike other inference…

3 11 29 3K 13

Download Image

Prashant @Prashant_1722

a year ago

Big Update on Google Gemma 2 Lmsys shows Gemma 2 is the king of open source AI LLM model 🔥🔥 > Beats Llama3 70B/ Qwen 72B/ Command R+ in LYMSYS Chat arena & 9B is the best < 15B model right now. > 2.5x smaller than Llama 3 & trained on 2/3rd the amount of tokens > Trained on…

0 1 3 424 0

Download Image

Google for Developers @googledevs

a year ago

@osanseviero ✈️ Love the tradition! 🤣 Gemma 2 is ready to take off! We're excited to see what our users build in the #GemmaVerse.

0 1 4 146 0

Omar Sanseviero @osanseviero

a year ago

As per my tradition of taking a flight when a new model comes out... 🥁🥁🥁 Welcome Google's Gemma 2: 9B and 27B models, with great initial results (benchmarks and lmsys) and friendly size for developers. Find models at Hugging Face huggingface.co/collections/go… More details soon!

2 18 128 6K 10

Omar Sanseviero @osanseviero

a year ago

Gemma 2 is out, and the @huggingface team wrote a blog post summarizing all the updates. Take a look 🤗 huggingface.co/blog/gemma2

1 31 135 8K 22

François Chollet @fchollet

a year ago

We're releasing a new high-performing open LLM: Gemma 2! Written in Keras 3, and compatible with TensorFlow, JAX, and PyTorch. Gemma 2 comes in two sizes, 9B and 27B parameters, with standard and instruction-tuned variants. Find them here: On Kaggle: kaggle.com/models/google/… On…

11 103 658 55K 165

Download Gif

ollama @ollama

a year ago

Google's Gemma 2 models are here! Available in 2 sizes: 9B: ollama run gemma2 27B: ollama run gemma2:27b 🗒️ you must have Ollama v0.1.47+ to run. Learn more: ollama.com/library/gemma2

21 128 637 61K 107

Download Image

Robert Dadashi @robdadashi

a year ago

I am so proud to announce that: - Gemma 2 27B IT tops all open weights models on Chatbot Arena, with a pinch of optimism in the face of uncertainty :) - Gemma 2 9B IT sets a new frontier for models of similar size. 1/n

4 24 126 23K 11

Download Image

Surya Bhupatiraju @suryabhupa

a year ago

I am absolutely thrilled to announce the release of Gemma 2! Today, we're releasing both pre-trained-only and fully post-trained 9B and 27B models. The full technical report is here: goo.gle/gemma2report and it's live *right now* on aistudio.google.com.

21 49 233 26K 50

Philipp Schmid @_philschmid

a year ago

Gemma 2 released! Google just released the next iteration of its open LLM! Gemma 2 comes in two sizes, 9B & 27B, trained on 13T tokens. Gemma 2 27B approaches @AIatMeta Llama 3 70B performance! First @lmsysorg Chatbot Arena evals place Gemma2 27B around @AnthropicAI Claude 3…

9 123 413 62K 121

Download Image

Wauplin @Wauplin

a year ago

You can now download Gemma models with Keras NLP directly from the Hugging Face Hub. Call that a nicely integrated ecosystem! Compatible models: huggingface.co/models?other=k…

0 5 14 3K 4

Download Image

Surya Bhupatiraju @suryabhupa

a year ago

Gemma v1.1 Instruct 2B and “7B” :) are out! See @robdadashi’s thread for details, featuring improvements in multi-turn, fixing some overly-chatty features, and new RL. Even more to come soon!

Robert Dadashi @robdadashi

a year ago

Gemma v1.1 Instruct 2B and “7B” :) are out! See @robdadashi’s thread for details, featuring improvements in multi-turn, fixing some overly-chatty features, and new RL. Even more to come soon!

13 68 367 324K 138

1 3 20 3K 3

Google for Developers @googledevs

a year ago

Say hello to the new members of the Gemma family of open models, tailored to devs and researchers. Learn how they offer new and innovative options for coding tasks and research exploration. 💬➡️ goo.gle/3TJGOHy

10 52 361 82K 50

Download Image

Samman e/lol @_gnikniht_

2 years ago

Today i have successfully fined tuned a nepali 2B gemma model. Special thanks to @ramsri_goutham for fine-tuning original model and also providing notebooks to train and inferences. You can use the model by clicking the link : colab.research.google.com/drive/1RJjaGhM…