Gemini 3.8 Flash posted scoring 73.7% on DeepSWE 1.1
doodlestein · x · 2026-09-03
Google's Logan Kilpatrick shared that Gemini 3.8 Flash scores 73.7% on the DeepSWE 1.1 coding benchmark. Developer @doodlestein noted that 3.7 Flash was already respectable and expressed excitement to try 3.8. This is an early score share with no full evaluation report yet.
More from Models
- Mocked benchmark shows DeepSeek's document chart grounding badly misaligned — burkov · 2026-10-18
- Dev complains Claude's habit of saying "not done" is his biggest pain point — jasonkneen · 2026-10-18
- Open-weight TYGER speech decision model makes 80ms turn-taking calls for full-duplex voice agents — solyarisoftware · 2026-10-18
- Paying Google AI subscriber blocked: custom-domain Workspace accounts can't access AI Studio — ChrisRackauckas · 2026-10-18
- Running Ling 3.1 Flash (560B MoE, 25B Active) Across 4 DGX Sparks via llama.cpp RPC — BerryBlush_Rogers691 · 2026-10-18
- Users press OpenAI on long-delayed Adult Mode a year after it was promised — ___Patrice___ · 2026-10-18