Zuckerberg: Meta already training post-Watermelon models on its 1GW Prometheus cluster
rohanpaul_ai · x · 2026-09-09
Mark Zuckerberg revealed on the Sources Podcast (via Alex Heath's channel) that Meta has moved into training post-Watermelon models.
- Watermelon is the unreleased successor to Avocado/Spark and is reportedly Meta's biggest model yet, trained with roughly 10X more compute.
- Quote: "We're using Prometheus (Meta's 1-gigawatt AI compute cluster) to now scale the post-watermelon models. Um we're past watermelon."
- Relayed by rohanpaulai with a link to the full video.
Related event: Zuckerberg Confirms Meta Training Post-Watermelon Larger Model(2 posts)→
More from Infra
- Encrypted GPT-2 speedrun challenges FHE inference: best is 3.031 s/token, goal sub-1s — tensorqt · 2026-10-27
- Open-source Latch compiles PyTorch into training megakernels, 2.75x faster on H100 — sguada · 2026-10-27
- NVIDIA claims HGX Rubin NVL8 delivers 10x tokens per megawatt vs B200 — thursdai_pod · 2026-10-27
- NVIDIA Claims HGX Rubin NVL8 Delivers 10x More Tokens per Megawatt Than B200 — thursdai_pod · 2026-10-27
- Together lets multiple LoRA post-training experiments share the same GPU capacity — togethercompute · 2026-10-27
- Commoditization dilemma: either human labor or GPUs become the commodity — _AustinCalvert_ · 2026-10-27