xAI completes training of 1.5 trillion parameter Grok V9
xAI completed training of its 1.5 trillion parameter Grok V9 model. The new system represents a substantial upgrade over the prior 0.5 trillion parameter V8 version. Training will continue with supplemental incorporation of Cursor data. This phase will be followed by supervised fine-tuning and reinforcement learning. xAI targets a public release of the model in three to four weeks.
We are improving the 0.5T Grok foundation model V8 (public version 4.3) every few days.
The 1.5T V9 just finished training (incorrectly called pre-training) and is a major upgrade. Next, we are adding the Cursor data in supplemental training (others call this mid-training), then SFT and RL. About 3 or 4 weeks to release.
This will be a banger.
#341Bill Yuchen Lin@BILLYUCHENLIN🚀🔜
Elon Musk@elonmuskWe are improving the 0.5T Grok foundation model V8 (public version 4.3) every few days. The 1.5T V9 just finished training (incorrectly called pre-training) and is a major upgrade. Next, we are adding the Cursor data in supplemental training (others call this mid-training), then SFT and RL. About 3 or 4 weeks to release. This will be a banger.
@elonmusk @orcdev oh fuck yes 🔥🔥
Elon Musk@elonmuskWe are improving the 0.5T Grok foundation model V8 (public version 4.3) every few days. The 1.5T V9 just finished training (incorrectly called pre-training) and is a major upgrade. Next, we are adding the Cursor data in supplemental training (others call this mid-training), then SFT and RL. About 3 or 4 weeks to release. This will be a banger.
@elonmusk @brivael @orcdev 🚀
Elon Musk@elonmuskWe are improving the 0.5T Grok foundation model V8 (public version 4.3) every few days. The 1.5T V9 just finished training (incorrectly called pre-training) and is a major upgrade. Next, we are adding the Cursor data in supplemental training (others call this mid-training), then SFT and RL. About 3 or 4 weeks to release. This will be a banger.
@elonmusk @orcdev any ETA?
Elon Musk@elonmuskWe are improving the 0.5T Grok foundation model V8 (public version 4.3) every few days. The 1.5T V9 just finished training (incorrectly called pre-training) and is a major upgrade. Next, we are adding the Cursor data in supplemental training (others call this mid-training), then SFT and RL. About 3 or 4 weeks to release. This will be a banger.