deepseek-v4.1POST

DeepSeek V4.1 Flash 1017 coding scores: Terminal-Bench 4.0 up to 42.3

The post cites Terminal-Bench 4.0 rising from 31.2 to 42.3 and DeepSWE from 74.2 to 77.0, plus 1M context, native vision and MIT weights.

INDIE | KAI

@adityaa_beheraArchived 11 days agoOriginally posted 2026.10.17

INDIE | KAI @adityaa_beheraOriginally posted 2026.10.17

Open original on X

DeepSeek V4.1-Flash-1017's coding jump is wild: • Terminal-Bench 4.0: 31.2 ➔ 42.3 • DeepSWE v1.1: 74.2 ➔ 77.0 • 1M context window + native vision • Faster runs & open weights (MIT) Crazy good for vibe coding on a budget. Anyone tested it in Cursor yet?

This post quotes another post: x.com/deepseek_ai/status/2111375640717033777

x.com/adityaa_behera/status/2111479429230989349Archived 2026.10.18 16:40