# DeepSeek V4.1 Flash's ARC-AGI-3 semi-private score more than doubles in new version

> Original title: Compared with the original DeepSeek V4.1 Flash (Sept 10, 2026), 1017 more than doubled its ARC-AGI-3 semi-private score: 2.68% → 5.81%. P...

- Source: ARC Prize
- Published: 2026-11-02T20:57:58.000Z
- Bad Influence: https://badinfluence.news/items/jmnmavwaa12dawjcclft9828q
- Original: https://x.com/arcprize/status/2117360590398079032

## Summary

ARC Prize says a new DeepSeek V4.1 Flash version, "1017," raised its ARC-AGI-3 semi-private score from 2.68% to 5.81% compared with the original model from September 10, 2026. Its public score fell from 11.05% to 8.97%, according to the post. Both runs used the provider adapter harness at max reasoning, with Baseten pricing of $0.30 per million input tokens and $1.20 per million output tokens.

## Post

Compared with the original DeepSeek V4.1 Flash (Sept 10, 2026), 1017 more than doubled its ARC-AGI-3 semi-private score: 2.68% → 5.81%. Public fell from 11.05% to 8.97%.

Both used the provider adapter harness at max reasoning with @Baseten pricing: $0.30/M input, $1.20/M output.

![](https://badinfluence.news/api/img-proxy?u=https%3A%2F%2Fpbs.twimg.com%2Fmedia%2FHWJgPBSWAAABaVK.jpg&mode=thumb&exp=1793923200&sig=ddea8e8c94f9b16d)