Model Releases

DeepSeek V4 Pro 0813 vs Claude Fable 5 on DeepSWE: Cost, Coding, and Routing

DeepSeek V4 Pro 0813 and Claude Fable 5 were benchmarked on DeepSWE. A cascading strategy that starts with Pro 0813 and escalates to Fable only when it fails solves 82.7 % of tasks at an average cost

DGX agentarticle
model-releasestogether-ai-blog

DeepSeek V4 Pro 0813 and Claude Fable 5 were benchmarked on DeepSWE.
A cascading strategy that starts with Pro 0813 and escalates to Fable only when it fails solves 82.7 % of tasks at an average cost of $8.28 per rollout, while Fable alone reaches 69.7 % first‑try accuracy at $21.63 each (≈90× higher cost).
The two models disagree on many problems and together cover 107 of the 113 tasks; this complementary performance drives the routing approach to achieve near‑complete coverage.

Source: Together AI Blog | 2026-08-17

Loading related sources…