Historical Fiction Spotlight: The Reflection in the Mirror (Murder in the Tower) by Gemma Morris-Conway tonyriches.blogspot.com/2025/12/hist... @gemma2.bsky.social #Tudors #HistoricalFiction
April 29, 2026 at 2:58 PM
Historical Fiction Spotlight: The Reflection in the Mirror (Murder in the Tower) by Gemma Morris-Conway tonyriches.blogspot.com/2025/12/hist... @gemma2.bsky.social #Tudors #HistoricalFiction
The road from zero to running a local model and submitting prompts feels friction-full.
With that said, I got the following stack working: ollama + gemma2 + GenKit.
With that said, I got the following stack working: ollama + gemma2 + GenKit.
December 17, 2024 at 4:30 AM
The road from zero to running a local model and submitting prompts feels friction-full.
With that said, I got the following stack working: ollama + gemma2 + GenKit.
With that said, I got the following stack working: ollama + gemma2 + GenKit.
Ovis1.6-Gemma2-27B was just released.
I used the 9B version in the past for document understanding and it was similar to Qwen2 VL 7B, so this one should be better (it's also explicitly mentioned "Enhanced Document Understanding" as a key feature).
huggingface.co/AIDC-AI/Ovis1.6-Gemma2-27B
I used the 9B version in the past for document understanding and it was similar to Qwen2 VL 7B, so this one should be better (it's also explicitly mentioned "Enhanced Document Understanding" as a key feature).
huggingface.co/AIDC-AI/Ovis1.6-Gemma2-27B
November 27, 2024 at 4:13 AM
Ovis1.6-Gemma2-27B was just released.
I used the 9B version in the past for document understanding and it was similar to Qwen2 VL 7B, so this one should be better (it's also explicitly mentioned "Enhanced Document Understanding" as a key feature).
huggingface.co/AIDC-AI/Ovis1.6-Gemma2-27B
I used the 9B version in the past for document understanding and it was similar to Qwen2 VL 7B, so this one should be better (it's also explicitly mentioned "Enhanced Document Understanding" as a key feature).
huggingface.co/AIDC-AI/Ovis1.6-Gemma2-27B
Update to this: I tried Gemma2:27b and I'm really impressed. It's the best model so far for my use case. Gemma coming in late but strong.
My open source LLM testing continues. Llama3 is still performing great 👍
I tested Mistral - it's not terrible but it's much worse than Llama3. Of the open source models in Ollama, which other ones are worth testing? 👀 👀
( MLSky DH CSS blueskAI 🤖 )
I tested Mistral - it's not terrible but it's much worse than Llama3. Of the open source models in Ollama, which other ones are worth testing? 👀 👀
( MLSky DH CSS blueskAI 🤖 )
July 22, 2024 at 5:24 PM
Update to this: I tried Gemma2:27b and I'm really impressed. It's the best model so far for my use case. Gemma coming in late but strong.
Quick Start Guide：Running Gemma2 on Glows.ai
Glows.ai provides high-performance, cost-effective AI cloud solutions, integrating advanced compute resources and tools to streamline development for users.
#DevOps #Cloud #glows.ai #konsttech #AIcomputing
Glows.ai provides high-performance, cost-effective AI cloud solutions, integrating advanced compute resources and tools to streamline development for users.
#DevOps #Cloud #glows.ai #konsttech #AIcomputing
Quick Start Guide：Running Gemma2 on Glows.ai
Gemma2 is a high-performance, open-source AI framework designed for developers and researchers. Known for its flexibility and ease of use…
medium.com
January 17, 2025 at 5:38 AM
Quick Start Guide：Running Gemma2 on Glows.ai
Glows.ai provides high-performance, cost-effective AI cloud solutions, integrating advanced compute resources and tools to streamline development for users.
#DevOps #Cloud #glows.ai #konsttech #AIcomputing
Glows.ai provides high-performance, cost-effective AI cloud solutions, integrating advanced compute resources and tools to streamline development for users.
#DevOps #Cloud #glows.ai #konsttech #AIcomputing
For what it's worth, it was kind of eye opening to try to turn an LLM into Trump and have it give more coherent hallucinatory answers than Trump's actual answers.
April 29, 2025 at 6:51 PM
For what it's worth, it was kind of eye opening to try to turn an LLM into Trump and have it give more coherent hallucinatory answers than Trump's actual answers.
at least this one does use LLM, it's Neta Lumina with gemma2 in place of clip
August 30, 2026 at 7:08 PM
at least this one does use LLM, it's Neta Lumina with gemma2 in place of clip
wamellow.com offers free #ai chat models in #discord, including various gemma2, llama3 and deepseek r1 models ;3
even if I have A's across the board in applied mathematics, it's far from a subject I appreciate or enjoy. For me #programming has nothing to do with maths
February 11, 2025 at 9:12 AM
wamellow.com offers free #ai chat models in #discord, including various gemma2, llama3 and deepseek r1 models ;3
Gemma2:9b running on mac m4-48gb, suuuuuper smoothly with pretty decent quality.
Integrated to obsidian in minutes.
Such a great project: ollama.com
Integrated to obsidian in minutes.
Such a great project: ollama.com
December 12, 2024 at 10:26 PM
Gemma2:9b running on mac m4-48gb, suuuuuper smoothly with pretty decent quality.
Integrated to obsidian in minutes.
Such a great project: ollama.com
Integrated to obsidian in minutes.
Such a great project: ollama.com
the AI, Gemma2, is a Google product that was fed lots and lots of things, many of whose authors disagree with it being used in AI.
you can't just count what it was *prompted* with
you can't just count what it was *prompted* with
December 23, 2024 at 3:06 PM
the AI, Gemma2, is a Google product that was fed lots and lots of things, many of whose authors disagree with it being used in AI.
you can't just count what it was *prompted* with
you can't just count what it was *prompted* with
Just done up another gemma2 project and quick demo.
It was pretty fun and all LOCAL
No API calls to a 'Strawberry model';
no data for ya 🍓
Happy New Year 🎇
It was pretty fun and all LOCAL
No API calls to a 'Strawberry model';
no data for ya 🍓
Happy New Year 🎇
January 1, 2025 at 10:49 AM
Just done up another gemma2 project and quick demo.
It was pretty fun and all LOCAL
No API calls to a 'Strawberry model';
no data for ya 🍓
Happy New Year 🎇
It was pretty fun and all LOCAL
No API calls to a 'Strawberry model';
no data for ya 🍓
Happy New Year 🎇
Quick Start Guide：Running Gemma2 on Glows.ai
Glows.ai provides high-performance, cost-effective AI cloud solutions, integrating advanced compute resources and tools to streamline development for users.
#DevOps #Cloud #glows.ai #konsttech #AIcomputing
Glows.ai provides high-performance, cost-effective AI cloud solutions, integrating advanced compute resources and tools to streamline development for users.
#DevOps #Cloud #glows.ai #konsttech #AIcomputing
Quick Start Guide：Running Gemma2 on Glows.ai
Gemma2 is a high-performance, open-source AI framework designed for developers and researchers. Known for its flexibility and ease of use…
medium.com
February 6, 2025 at 12:47 PM
Quick Start Guide：Running Gemma2 on Glows.ai
Glows.ai provides high-performance, cost-effective AI cloud solutions, integrating advanced compute resources and tools to streamline development for users.
#DevOps #Cloud #glows.ai #konsttech #AIcomputing
Glows.ai provides high-performance, cost-effective AI cloud solutions, integrating advanced compute resources and tools to streamline development for users.
#DevOps #Cloud #glows.ai #konsttech #AIcomputing
Oh thank you for finding this message! Gemma2 is basically the same as ChatGPT, a massive copyright infringement machine that requires the energy of the sun to do anything. Worst part is that he probably doesn't even know it (see my reply thread to Breadsticc)
December 23, 2024 at 7:42 AM
Oh thank you for finding this message! Gemma2 is basically the same as ChatGPT, a massive copyright infringement machine that requires the energy of the sun to do anything. Worst part is that he probably doesn't even know it (see my reply thread to Breadsticc)
ui <- bslib::page_fluid(chat_ui("chat"))
server <- function(input, output, session) {
chat <- ellmer::chat_ollama(model = "gemma2")
observeEvent(input$chat_user_input, {
stream <- chat$stream_async(input$chat_user_input)
chat_append("chat", stream)
})
}
shinyApp(ui, server)
2/2 #RStats #GenAI
server <- function(input, output, session) {
chat <- ellmer::chat_ollama(model = "gemma2")
observeEvent(input$chat_user_input, {
stream <- chat$stream_async(input$chat_user_input)
chat_append("chat", stream)
})
}
shinyApp(ui, server)
2/2 #RStats #GenAI
January 30, 2025 at 4:10 PM
"gems, Gemma, gemma2,paligemma,gem9"
#GoogleIO
#GoogleIO
What Is This Whats Happening GIF
ALT: What Is This Whats Happening GIF
media.tenor.com
May 15, 2024 at 6:09 AM
"gems, Gemma, gemma2,paligemma,gem9"
#GoogleIO
#GoogleIO
My Saturday #buildinpublic
I am doing a #Gemma2 > Android integration.
Pretty happy with a 2 day part time effort!
Sweet points:
• Ollama
• Open #LLM
• 9B parameter size
• Cloud GPU acceleration
Nicer GUI next!
I can see the privacy use cases 🕵️ since we control the whole API
#Gemmaverse
I am doing a #Gemma2 > Android integration.
Pretty happy with a 2 day part time effort!
Sweet points:
• Ollama
• Open #LLM
• 9B parameter size
• Cloud GPU acceleration
Nicer GUI next!
I can see the privacy use cases 🕵️ since we control the whole API
#Gemmaverse
January 4, 2025 at 10:42 AM
My Saturday #buildinpublic
I am doing a #Gemma2 > Android integration.
Pretty happy with a 2 day part time effort!
Sweet points:
• Ollama
• Open #LLM
• 9B parameter size
• Cloud GPU acceleration
Nicer GUI next!
I can see the privacy use cases 🕵️ since we control the whole API
#Gemmaverse
I am doing a #Gemma2 > Android integration.
Pretty happy with a 2 day part time effort!
Sweet points:
• Ollama
• Open #LLM
• 9B parameter size
• Cloud GPU acceleration
Nicer GUI next!
I can see the privacy use cases 🕵️ since we control the whole API
#Gemmaverse
⚔️⏱️New TLI season - fastest VTuber in Torchlight speeding to bosses - Freeze Gemma2 melee
#VTuber #VTuberUprising
www.twitch.tv/megami_infini
#VTuber #VTuberUprising
www.twitch.tv/megami_infini
Twitch
Twitch is the world
www.twitch.tv
October 25, 2024 at 1:44 AM
⚔️⏱️New TLI season - fastest VTuber in Torchlight speeding to bosses - Freeze Gemma2 melee
#VTuber #VTuberUprising
www.twitch.tv/megami_infini
#VTuber #VTuberUprising
www.twitch.tv/megami_infini
The new bggpt is here. Based on Gemma2. The large 27B model is on par with gpt4o with gpt4o used as a judge.
models.bggpt.ai/blog/
models.bggpt.ai/blog/
State-of-the-art Bulgarian LLMs
State-of-the-art generative AI created for the Bulgarian government, users, public and private organizations
models.bggpt.ai
November 19, 2024 at 12:38 PM
The new bggpt is here. Based on Gemma2. The large 27B model is on par with gpt4o with gpt4o used as a judge.
models.bggpt.ai/blog/
models.bggpt.ai/blog/
Gemma has a goal of 50 boxes getting donated! Even one box makes a difference 💚🍪
digitalcookie.girlscouts.org/scout/gemma2...
digitalcookie.girlscouts.org/scout/gemma2...
February 27, 2025 at 12:23 PM
Gemma has a goal of 50 boxes getting donated! Even one box makes a difference 💚🍪
digitalcookie.girlscouts.org/scout/gemma2...
digitalcookie.girlscouts.org/scout/gemma2...
To clarify, they're not using AI to generate sprites, they're generating Pokédex entries in text. Two reasons why this is shitty:
1) the language model they use (gemma2) is trained using massive amounts of stolen data
2) they're refusing artists who want their contributions removed from the game
1) the language model they use (gemma2) is trained using massive amounts of stolen data
2) they're refusing artists who want their contributions removed from the game
December 23, 2024 at 8:25 AM
To clarify, they're not using AI to generate sprites, they're generating Pokédex entries in text. Two reasons why this is shitty:
1) the language model they use (gemma2) is trained using massive amounts of stolen data
2) they're refusing artists who want their contributions removed from the game
1) the language model they use (gemma2) is trained using massive amounts of stolen data
2) they're refusing artists who want their contributions removed from the game
New release of torchtune right before Christmas! We have new recipes, better integration with vLLM and HF, support for Gemma2, and more. We've also now added support for Kaggle notebooks!
www.kaggle.com/code/felipem...
www.kaggle.com/code/felipem...
torchtune in kaggle
Explore and run machine learning code with Kaggle Notebooks | Using data from multiple data sources
www.kaggle.com
December 20, 2024 at 9:54 PM
New release of torchtune right before Christmas! We have new recipes, better integration with vLLM and HF, support for Gemma2, and more. We've also now added support for Kaggle notebooks!
www.kaggle.com/code/felipem...
www.kaggle.com/code/felipem...
Assuming you got at least 3 of these, easiest way to get going is grab ollama:
ollama.com
Then:
ollama pull gemma2:27b-instruct-q4_K_M
ollama run gemma2:27b-instruct-q4_K_M
This will drop you into an interactive chat session with a "state of the art" local model
ollama.com
Then:
ollama pull gemma2:27b-instruct-q4_K_M
ollama run gemma2:27b-instruct-q4_K_M
This will drop you into an interactive chat session with a "state of the art" local model
gemma2
Google Gemma 2 is a high-performing and efficient model available in three sizes: 2B, 9B, and 27B.
ollama.com
October 13, 2024 at 8:51 PM
Assuming you got at least 3 of these, easiest way to get going is grab ollama:
ollama.com
Then:
ollama pull gemma2:27b-instruct-q4_K_M
ollama run gemma2:27b-instruct-q4_K_M
This will drop you into an interactive chat session with a "state of the art" local model
ollama.com
Then:
ollama pull gemma2:27b-instruct-q4_K_M
ollama run gemma2:27b-instruct-q4_K_M
This will drop you into an interactive chat session with a "state of the art" local model