Anthropic's model reportedly finished training in October and cleared 50% on every benchmark it was tested on.
One account describes Model 4 as the highest-scoring model in the set METR analyzed. Another post claims it scored 178.6 on Anthropic ECI, 9.24 points above Opus 5.5, a gap the poster equates to roughly 7.5 months of frontier progress.
—
5 Sources, first seen ago
The departing team member says Claude Code now handles large migrations and builds new products.
FlashProve aims to verify replies against a provider’s committed model without revealing weights, but proof generation takes substantial GPU time.
The departing team member says Claude Code now handles large migrations and builds new products.
FlashProve aims to verify replies against a provider’s committed model without revealing weights, but proof generation takes substantial GPU time.
#8
#8
@scaling01HOLY FUCK FINALLY some information on Anthropic internal models: Model 4 finished training in October 2026 and scores 178.6 on the Anthropic ECI this is 9.24 ECI higher than Opus 5.5 or ~7.5 months of frontier progress
@AndrewCurran_Anthropic has a new internal model that finished training last month. METR refers to it as Model 4 (Anthropic has its own internal codename). It is the highest-scoring model in the set METR analyzed, and it is the first to clear 50% on every benchmark it was tested on.@scaling01HOLY FUCK FINALLY some information on Anthropic internal models: Model 4 finished training in October 2026 and scores 178.6 on the Anthropic ECI this is 9.24 ECI higher than Opus 5.5 or ~7.5 months of frontier progress@AndrewCurran_Anthropic has a new internal model that finished training last month. METR refers to it as Model 4 (Anthropic has its own internal codename). It is the highest-scoring model in the set METR analyzed, and it is the first to clear 50% on every benchmark it was tested on.@scaling01HOLY FUCK FINALLY some information on Anthropic internal models: Model 4 finished training in October 2026 and scores 178.6 on the Anthropic ECI this is 9.24 ECI higher than Opus 5.5 or ~7.5 months of frontier progress@AndrewCurran_Anthropic has a new internal model that finished training last month. METR refers to it as Model 4 (Anthropic has its own internal codename). It is the highest-scoring model in the set METR analyzed, and it is the first to clear 50% on every benchmark it was tested on.@scaling01HOLY FUCK FINALLY some information on Anthropic internal models: Model 4 finished training in October 2026 and scores 178.6 on the Anthropic ECI this is 9.24 ECI higher than Opus 5.5 or ~7.5 months of frontier progress@AndrewCurran_Anthropic has a new internal model that finished training last month. METR refers to it as Model 4 (Anthropic has its own internal codename). It is the highest-scoring model in the set METR analyzed, and it is the first to clear 50% on every benchmark it was tested on.
