The crux of it all…
Gift article: www.theatlantic.com/politics/arc...
Gift article: www.theatlantic.com/politics/arc...
September 9, 2025 at 1:07 AM
The crux of it all…
Gift article: www.theatlantic.com/politics/arc...
Gift article: www.theatlantic.com/politics/arc...
Como eu pude me esquecer do Chuck Bartowski.
August 15, 2025 at 2:10 AM
Como eu pude me esquecer do Chuck Bartowski.
Comment Chuck? Chuck Bartowski ?
August 7, 2025 at 7:17 PM
Comment Chuck? Chuck Bartowski ?
There are news of new reasoning models from nvidia.
I downloaded the Q4_K_L quantization (as usual) from Bartowski (good past experience) and this is the first time I experienced a broken model/quantization:
$ llm 'Why did the chicken cross the road?'
of * if! (true,,)
#llm
I downloaded the Q4_K_L quantization (as usual) from Bartowski (good past experience) and this is the first time I experienced a broken model/quantization:
$ llm 'Why did the chicken cross the road?'
of * if! (true,,)
#llm
July 26, 2025 at 7:08 AM
There are news of new reasoning models from nvidia.
I downloaded the Q4_K_L quantization (as usual) from Bartowski (good past experience) and this is the first time I experienced a broken model/quantization:
$ llm 'Why did the chicken cross the road?'
of * if! (true,,)
#llm
I downloaded the Q4_K_L quantization (as usual) from Bartowski (good past experience) and this is the first time I experienced a broken model/quantization:
$ llm 'Why did the chicken cross the road?'
of * if! (true,,)
#llm
🎙️ Join our 🔴𝐓𝐞𝐜𝐡 𝐁𝐞𝐚𝐭𝐬 𝐋𝐢𝐯𝐞 Show!
🗓️ Thursday 17th 11:30 AM EDT
🎯 A chill livestream unpacking LLM #Quantization: #vllm vs #ollama. Learn about the What & How.
🔥Dope guest stars:
#bartowski from arcee.ai & Eldar Kurtic from #RedHat
🔗Stream on YouTube & Linkedin:
www.youtube.com/watch?v=XTE0...
🗓️ Thursday 17th 11:30 AM EDT
🎯 A chill livestream unpacking LLM #Quantization: #vllm vs #ollama. Learn about the What & How.
🔥Dope guest stars:
#bartowski from arcee.ai & Eldar Kurtic from #RedHat
🔗Stream on YouTube & Linkedin:
www.youtube.com/watch?v=XTE0...
🔴TechBeats live : LLM Quantization "vLLM vs. Ollama"
YouTube video by CloudDude
www.youtube.com
July 10, 2025 at 2:33 PM
🎙️ Join our 🔴𝐓𝐞𝐜𝐡 𝐁𝐞𝐚𝐭𝐬 𝐋𝐢𝐯𝐞 Show!
🗓️ Thursday 17th 11:30 AM EDT
🎯 A chill livestream unpacking LLM #Quantization: #vllm vs #ollama. Learn about the What & How.
🔥Dope guest stars:
#bartowski from arcee.ai & Eldar Kurtic from #RedHat
🔗Stream on YouTube & Linkedin:
www.youtube.com/watch?v=XTE0...
🗓️ Thursday 17th 11:30 AM EDT
🎯 A chill livestream unpacking LLM #Quantization: #vllm vs #ollama. Learn about the What & How.
🔥Dope guest stars:
#bartowski from arcee.ai & Eldar Kurtic from #RedHat
🔗Stream on YouTube & Linkedin:
www.youtube.com/watch?v=XTE0...
I saw daddy bartowski and was summoned so quickly hahahaha hi ebee!!! I do not watch Star Trek, I’m so sorry but I do like the actor who plays chucks dad?!?!? So no, not sick!!!
June 9, 2025 at 8:15 PM
I saw daddy bartowski and was summoned so quickly hahahaha hi ebee!!! I do not watch Star Trek, I’m so sorry but I do like the actor who plays chucks dad?!?!? So no, not sick!!!
Yessssss I love Chuck and Daddy Bartowski can get ittttttt
June 9, 2025 at 8:13 PM
Yessssss I love Chuck and Daddy Bartowski can get ittttttt
Did you ever watch chuck? Bakula as Daddy bartowski does something to me. Is @rapunzerelli.bsky.social on here? Jojo am I sick for this
June 9, 2025 at 8:12 PM
Did you ever watch chuck? Bakula as Daddy bartowski does something to me. Is @rapunzerelli.bsky.social on here? Jojo am I sick for this
We're playing #GuessTheMovie on the #WorkdayWinddown. But can you guess it from this cryptic clue?
"Chuck Bartowski becomes a Superhero.."
"Chuck Bartowski becomes a Superhero.."
June 6, 2025 at 4:24 PM
We're playing #GuessTheMovie on the #WorkdayWinddown. But can you guess it from this cryptic clue?
"Chuck Bartowski becomes a Superhero.."
"Chuck Bartowski becomes a Superhero.."
OpenThinker3 is out. It's a 7B model that beats every open 7B/8B model on math, code, and science - including RL-trained ones. To run it locally, click Use this model on @huggingface and select 👋 Jan.
huggingface.co/bartowski/o...
huggingface.co/bartowski/o...
bartowski/open-thoughts_OpenThinker3-7B-GGUF · Hugging Face
huggingface.co
June 6, 2025 at 12:52 AM
OpenThinker3 is out. It's a 7B model that beats every open 7B/8B model on math, code, and science - including RL-trained ones. To run it locally, click Use this model on @huggingface and select 👋 Jan.
huggingface.co/bartowski/o...
huggingface.co/bartowski/o...
rip chuck bartowski you would have hated zachary lehi
a man in a white shirt and tie is looking at the camera in a store .
ALT: a man in a white shirt and tie is looking at the camera in a store .
media.tenor.com
May 28, 2025 at 2:11 AM
rip chuck bartowski you would have hated zachary lehi
My man looks like bizzaro Vaudevillian Chuck Bartowski
May 20, 2025 at 3:03 AM
My man looks like bizzaro Vaudevillian Chuck Bartowski
An example of a startup command
llama-server.exe -hf bartowski/google_gemma-3-12b-it-qat-GGUF -c 40960
Then, access 127.0.0.1
Since it is a long text, the context length is increased with -c
llama-server.exe -hf bartowski/google_gemma-3-12b-it-qat-GGUF -c 40960
Then, access 127.0.0.1
Since it is a long text, the context length is increased with -c
May 16, 2025 at 5:48 AM
An example of a startup command
llama-server.exe -hf bartowski/google_gemma-3-12b-it-qat-GGUF -c 40960
Then, access 127.0.0.1
Since it is a long text, the context length is increased with -c
llama-server.exe -hf bartowski/google_gemma-3-12b-it-qat-GGUF -c 40960
Then, access 127.0.0.1
Since it is a long text, the context length is increased with -c
GGUF models on Hugging Face run in Jan with one click.
Click "Use this model", select 👋 Jan.
Thanks to Hugging Face folks, GG's llama.cpp and the quant crew; Bartowski, Unsloth, and others.
Click "Use this model", select 👋 Jan.
Thanks to Hugging Face folks, GG's llama.cpp and the quant crew; Bartowski, Unsloth, and others.
May 13, 2025 at 6:20 AM
GGUF models on Hugging Face run in Jan with one click.
Click "Use this model", select 👋 Jan.
Thanks to Hugging Face folks, GG's llama.cpp and the quant crew; Bartowski, Unsloth, and others.
Click "Use this model", select 👋 Jan.
Thanks to Hugging Face folks, GG's llama.cpp and the quant crew; Bartowski, Unsloth, and others.
Somehow Qwen3 30B is 30% faster than Gemma 3 27B. Both Bartowski GGUF static quants. Gemma 3 at q5_k_l, Qwen3 at Q6_k_l. What the heck? How did they pull that off? Its reasoning is really good too.
April 29, 2025 at 12:20 AM
Somehow Qwen3 30B is 30% faster than Gemma 3 27B. Both Bartowski GGUF static quants. Gemma 3 at q5_k_l, Qwen3 at Q6_k_l. What the heck? How did they pull that off? Its reasoning is really good too.
Já faz um mês que estou vendo os vídeos do Rapha Bartowski no YouTube onde a galera manda vídeo descolorindo ou cabelo ou pintando de ruivo/vermelho, isso me tentou a voltar a pintar o cabelo de preto nem que seja com aquelas máscaras pigmentantes, tipo Viúva Negra da Kamaleão Color
April 25, 2025 at 12:47 PM
Já faz um mês que estou vendo os vídeos do Rapha Bartowski no YouTube onde a galera manda vídeo descolorindo ou cabelo ou pintando de ruivo/vermelho, isso me tentou a voltar a pintar o cabelo de preto nem que seja com aquelas máscaras pigmentantes, tipo Viúva Negra da Kamaleão Color
The Great Chuck Bartowski - IYKYK
a man in a white shirt and tie stands in front of a buy sign
ALT: a man in a white shirt and tie stands in front of a buy sign
media.tenor.com
April 23, 2025 at 6:45 PM
The Great Chuck Bartowski - IYKYK
Bartowski quantized ReZero - our 3B model that learns to search.
Big thanks to Bartowski for jumping on it early and releasing the GGUF version: huggingface.co/bartowski/M...
Big thanks to Bartowski for jumping on it early and releasing the GGUF version: huggingface.co/bartowski/M...
bartowski/Menlo_ReZero-v0.1-llama-3.2-3b-it-grpo-250404-GGUF · Hugging Face
huggingface.co
April 17, 2025 at 1:30 AM
Bartowski quantized ReZero - our 3B model that learns to search.
Big thanks to Bartowski for jumping on it early and releasing the GGUF version: huggingface.co/bartowski/M...
Big thanks to Bartowski for jumping on it early and releasing the GGUF version: huggingface.co/bartowski/M...
Instead of optimizing for recall or speed, we train the model to retry when it's wrong - using reinforcement learning to build persistence into the search process.
Thanks to Meta for the Llama 3.2B base, Unsloth for AutoDidact (the framework we built on), and Bartowski for quantizing the model!
Thanks to Meta for the Llama 3.2B base, Unsloth for AutoDidact (the framework we built on), and Bartowski for quantizing the model!
April 16, 2025 at 8:55 AM
Instead of optimizing for recall or speed, we train the model to retry when it's wrong - using reinforcement learning to build persistence into the search process.
Thanks to Meta for the Llama 3.2B base, Unsloth for AutoDidact (the framework we built on), and Bartowski for quantizing the model!
Thanks to Meta for the Llama 3.2B base, Unsloth for AutoDidact (the framework we built on), and Bartowski for quantizing the model!
Llama-3.1-8B-UltraLong: 4M context — that's 30× longer than the base model.
It enables full-document, multi-document, or long-reasoning tasks with no chunking.
To run it locally: click "Use this model" on Hugging Face and select Jan: huggingface.co/bartowski/n...
It enables full-document, multi-document, or long-reasoning tasks with no chunking.
To run it locally: click "Use this model" on Hugging Face and select Jan: huggingface.co/bartowski/n...
bartowski/nvidia_Llama-3.1-8B-UltraLong-4M-Instruct-GGUF · Hugging Face
huggingface.co
April 14, 2025 at 2:37 AM
Llama-3.1-8B-UltraLong: 4M context — that's 30× longer than the base model.
It enables full-document, multi-document, or long-reasoning tasks with no chunking.
To run it locally: click "Use this model" on Hugging Face and select Jan: huggingface.co/bartowski/n...
It enables full-document, multi-document, or long-reasoning tasks with no chunking.
To run it locally: click "Use this model" on Hugging Face and select Jan: huggingface.co/bartowski/n...
ZR1-1.5B: A 1.5B reasoning model that outperforms Llama-3.1-70B on hard coding.
Shows 50%+ improvement over the base R1-Distill-1.5B, achieving 37.91% pass@1 on GPQA-Diamond.
To run it locally: click "Use this model"on @huggingface and select Jan. huggingface.co/bartowski/Z...
Shows 50%+ improvement over the base R1-Distill-1.5B, achieving 37.91% pass@1 on GPQA-Diamond.
To run it locally: click "Use this model"on @huggingface and select Jan. huggingface.co/bartowski/Z...
bartowski/Zyphra_ZR1-1.5B-GGUF · Hugging Face
huggingface.co
April 11, 2025 at 4:34 AM
ZR1-1.5B: A 1.5B reasoning model that outperforms Llama-3.1-70B on hard coding.
Shows 50%+ improvement over the base R1-Distill-1.5B, achieving 37.91% pass@1 on GPQA-Diamond.
To run it locally: click "Use this model"on @huggingface and select Jan. huggingface.co/bartowski/Z...
Shows 50%+ improvement over the base R1-Distill-1.5B, achieving 37.91% pass@1 on GPQA-Diamond.
To run it locally: click "Use this model"on @huggingface and select Jan. huggingface.co/bartowski/Z...
EXAONE-Deep-32B: 32B model outperforms 70B DeepSeek-R1-Llama on AIME 2025 (65.8% vs 53.9%).
Achieves 66.1% on GPQA Diamond vs 65.2% for DeepSeek-R1-Llama-70B.
To run it locally, click Use this model on Hugging Face and select Jan: huggingface.co/bartowski/L...
Achieves 66.1% on GPQA Diamond vs 65.2% for DeepSeek-R1-Llama-70B.
To run it locally, click Use this model on Hugging Face and select Jan: huggingface.co/bartowski/L...
bartowski/LGAI-EXAONE_EXAONE-Deep-32B-GGUF · Hugging Face
huggingface.co
April 9, 2025 at 3:15 AM
EXAONE-Deep-32B: 32B model outperforms 70B DeepSeek-R1-Llama on AIME 2025 (65.8% vs 53.9%).
Achieves 66.1% on GPQA Diamond vs 65.2% for DeepSeek-R1-Llama-70B.
To run it locally, click Use this model on Hugging Face and select Jan: huggingface.co/bartowski/L...
Achieves 66.1% on GPQA Diamond vs 65.2% for DeepSeek-R1-Llama-70B.
To run it locally, click Use this model on Hugging Face and select Jan: huggingface.co/bartowski/L...
GemmaCoder-12B: Code-specialized Gemma-12B boosting LiveCodeBench by 50% (21.9% → 32.9%).
Fine-tuned via SFT on competitive coding (Codeforces). Thanks @ben_burtenshaw!
To run it locally, click Use this model on @huggingface and select Jan: huggingface.co/bartowski/b...
Fine-tuned via SFT on competitive coding (Codeforces). Thanks @ben_burtenshaw!
To run it locally, click Use this model on @huggingface and select Jan: huggingface.co/bartowski/b...
bartowski/burtenshaw_GemmaCoder3-12B-GGUF · Hugging Face
huggingface.co
April 8, 2025 at 2:33 AM
GemmaCoder-12B: Code-specialized Gemma-12B boosting LiveCodeBench by 50% (21.9% → 32.9%).
Fine-tuned via SFT on competitive coding (Codeforces). Thanks @ben_burtenshaw!
To run it locally, click Use this model on @huggingface and select Jan: huggingface.co/bartowski/b...
Fine-tuned via SFT on competitive coding (Codeforces). Thanks @ben_burtenshaw!
To run it locally, click Use this model on @huggingface and select Jan: huggingface.co/bartowski/b...