Grok 4.5 Training Progress Revealed
Arindam_1729 · x · 2026-07-18
The context of this post mentions:
- The author is currently testing the Grok model and subjectively评价 it as "much better than Fable."
- A quoted reply states that xAI's 2T model is "better in every way" than the 1.5T version, with initial training expected to finish next week.
- It also mentions that it is expected to surpass Kimi, while maintaining speeds and token efficiency close to the 1.5T version, which is Grok 4.5.
Overall, this conveys the training progress and performance expectations for the new generation of Grok models.
More from Models
- Benchmark scores drop from 89% to 19% on new evals — how benchmaxxing breaks leaderboard trust — airesearch12 · 2026-09-11
- ChatGPT tells user their question is too hard and to 'accept dumber answers' — phido3000 · 2026-09-11
- Developer Building a Unified Leaderboard of All Model Benchmark Scores — airesearch12 · 2026-09-11
- Rumor claims Kimi faked performance by serving Claude; DeepSeek new model surprises in evals — realsohamparekh · 2026-09-11
- GPT-5.6 writes well but is instantly forgettable, user complains — BasedRaddka · 2026-09-11
- Opus Refuses Protein Research Codebase Over 'Safety' Concerns, Dev Considers Rolling His Own — josephdviviano · 2026-09-11