Mar 29, 04:47 AM

xAI Unveils Grok-1.5 with Enhanced Reasoning, 128K Token Context, MATH 50.6%, GSM8K 90%

xAI has announced the release of Grok-1.5, an upgraded large language model with improved reasoning capabilities and a context length of 128,000 tokens. Grok-1.5 achieved a 50.6% score on the MATH benchmark and a 90% score on the GSM8K benchmark, showcasing its performance in math and coding tasks. The model is built on a custom distributed training framework based on JAX, Rust, and Kubernetes, and will soon be available to early testers and existing users on social media platform X.

#xAI #Grok #JAX #Rust #Kubernetes

Written with ChatGPT (GPT-3).

Sources

Marktechpost AI Dev News ⚡@Marktechpost
3 years ago
https://t.co/5mtsSzkpqN Announces Grok 1.5: A Look at the Improved Reasoning and Long Context Capabilities https://t.co/OXeIVBynyt @X @xai #ArtificialIntelligence
proper@ProperPrompter
3 years ago
Hopefully the use cases for Grok 1.5 extend beyond just roasting yourself. 😀
TuringPost@TheTuringPost
3 years ago
xAI announced Grok-1.5 with a 128,000-token context window. Improvements in performance benchmarks: - MATH: 50.6% - GSM8K: 90% - HumanEval: 74.1% 1/4 https://t.co/3VClFdn16v

Additional media

Image #1 for story xai-unveils-grok-1-5-enhanced-reasoning-128k-token-context-6

Image #2 for story xai-unveils-grok-1-5-enhanced-reasoning-128k-token-context-6

Image #3 for story xai-unveils-grok-1-5-enhanced-reasoning-128k-token-context-6

Image #4 for story xai-unveils-grok-1-5-enhanced-reasoning-128k-token-context-6

Image #5 for story xai-unveils-grok-1-5-enhanced-reasoning-128k-token-context-6