Snorlax/ppo-Pyramids
Reinforcement Learning
•
Updated
•
19
Shangzhi Zhang
Snorlax
AI & ML interests
None yet
Recent Activity
updated
a model
1 day ago
Snorlax/ppo-Pyramids
updated
a model
1 day ago
Snorlax/ppo-SnowballTarget
updated
a model
1 day ago
Snorlax/Reinforce-PixelCopter
Organizations
Snorlax's activity
updated
3
models
1 day ago
updated
2
models
2 days ago
updated
2
models
7 days ago
liked
a
Space
7 days ago
updated
a
model
7 days ago
updated
a
model
8 days ago
liked
a
Space
4 months ago
liked
a
model
4 months ago
upvoted
an
article
5 months ago
Article
SmolLM - blazingly fast and remarkably powerful
•
279
liked
a
Space
6 months ago
Potential bug when batch inferencing with left padding.
1
#26 opened 6 months ago
by
Snorlax
Snorlax
liked
a
dataset
7 months ago
reacted to
georgewritescode's
post with 👍
7 months ago
Post
1006
Visualization of GPT-4o breaking away from the quality & speed trade-off curve the LLMs have followed thus far ✂️
Key GPT-4o takeaways
‣ GPT-4o not only offers the highest quality, it also sits amongst the fastest LLMs
‣ For those with speed/latency-sensitive use cases, where previously Claude 3 Haiku or Mixtral 8x7b were leaders, GPT-4o is now a compelling option (though significantly more expensive)
‣ Previously Groq was the only provider to break from the curve using its own LPU chips. OpenAI has done it on Nvidia hardware (one can imagine the potential for GPT-4o on Groq)
👉 How did they do it? Will follow up with more analysis on this but potential approaches include a very large but sparse MoE model (similar to Snowflake's Arctic) and improvements in data quality (likely to have driven much of Llama 3's impressive quality relative to parameter count)
Notes: Throughput represents the median across providers over the last 14 days of measurements (8x per day)
Data is present on our HF leaderboard: ArtificialAnalysis/LLM-Performance-Leaderboard and graphs present on our website
Key GPT-4o takeaways
‣ GPT-4o not only offers the highest quality, it also sits amongst the fastest LLMs
‣ For those with speed/latency-sensitive use cases, where previously Claude 3 Haiku or Mixtral 8x7b were leaders, GPT-4o is now a compelling option (though significantly more expensive)
‣ Previously Groq was the only provider to break from the curve using its own LPU chips. OpenAI has done it on Nvidia hardware (one can imagine the potential for GPT-4o on Groq)
👉 How did they do it? Will follow up with more analysis on this but potential approaches include a very large but sparse MoE model (similar to Snowflake's Arctic) and improvements in data quality (likely to have driven much of Llama 3's impressive quality relative to parameter count)
Notes: Throughput represents the median across providers over the last 14 days of measurements (8x per day)
Data is present on our HF leaderboard: ArtificialAnalysis/LLM-Performance-Leaderboard and graphs present on our website
- 1 reply
updated
a
collection
9 months ago
updated
a
collection
10 months ago