DEEPSEEK · MODEL RELEASE TRACKER

DeepSeek-V4.1-Flash

DeepSeek announced DeepSeek-V4.1-Flash on 2026-09-10 as a smaller, faster model in its new architecture family with native visual understanding, optimized KV cache, and improved cost/performance versus earlier V4 models.

UPDATED 2026-09-30

CURRENT SNAPSHOT3/5 DIMENSIONS WITH DATA

The dimensions that change the decision.

PRICING

Not established from the available sources.

CONTEXT WINDOW

Not established from the available sources.

MODALITIES

ModalityNative visual understanding (vision-capable, multimodal).2026-09-10

BENCHMARKS

Benchmark comparison vs flagship modelsTests by multiple parties and internal evaluation place V4.1-Flash ahead of DeepSeek-V4-Pro on performance, cost, speed, and total runtime.2026-09-10

AVAILABILITY

Retirement of older variants and compatibility routingV4-Flash and V4-Flash-Vision-Exp are retired; requests to deepseek-v4-flash and deepseek-v4-flash-vision-exp temporarily route to V4.1-Flash for compatibility.2026-09-10

V4-Pro routing scheduleStarting at 04:00 UTC on 2026-09-14, all deepseek-v4-pro requests will route to V4.1-Flash at V4.1-Flash rates until V4.1-Pro launches.2026-09-10

Official partner supportOfficial partners WorkBuddy (including CodeBuddy) and OpenCode fully support V4.1-Flash.2026-09-10

RELEASE TIMELINE

Published, source-backed release events only.

2026-09-10

1 SOURCES · IMPORTANCE 8.0

DeepSeek launches DeepSeek‑V4.1‑Flash with new causal encoder–decoder and lower costs

DeepSeek announced DeepSeek‑V4.1‑Flash, a smaller-native-vision model using a new Causal Encoder–Decoder design (8B active params for input, 16B for output), new pretraining and larger-scale RL post-training, plus KV‑cache compression. The company is retiring V4‑Flash variants, routing deepseek-v4-pro traffic to V4.1‑Flash from 04:00 UTC on Sept 14, 2026, and changing pricing (new rates take effect 04:00 UTC on Sept 10, 2026) while partners WorkBuddy, CodeBuddy and OpenCode already support V4.1‑Flash.

→

VERIFIABLE FACTS

Every value stays attached to a source and date.

RELEASE · Release announcementDEVELOPER CLAIM

DeepSeek announced DeepSeek-V4.1-Flash on 2026-09-10 as the smallest model in its new architecture family with native visual understanding.

CAPABILITIES · Architecture and active parameter countsDEVELOPER CLAIM

Causal Encoder–Decoder architecture with approximately 8B active parameters for input and 16B for output.

MODALITIES · ModalityDEVELOPER CLAIM

Native visual understanding (vision-capable, multimodal).

CAPABILITIES · KV cache requirements and cost efficiencyDEVELOPER CLAIM

Compared with the previous generation, V4.1-Flash reduces KV cache needs via compression, significantly cutting cache-hit charges and agent costs.

BENCHMARKS · Benchmark comparison vs flagship modelsDEVELOPER CLAIM

Tests by multiple parties and internal evaluation place V4.1-Flash ahead of DeepSeek-V4-Pro on performance, cost, speed, and total runtime.

AVAILABILITY · Retirement of older variants and compatibility routingDEVELOPER CLAIM

V4-Flash and V4-Flash-Vision-Exp are retired; requests to deepseek-v4-flash and deepseek-v4-flash-vision-exp temporarily route to V4.1-Flash for compatibility.

AVAILABILITY · V4-Pro routing scheduleDEVELOPER CLAIM

Starting at 04:00 UTC on 2026-09-14, all deepseek-v4-pro requests will route to V4.1-Flash at V4.1-Flash rates until V4.1-Pro launches.

AVAILABILITY · Official partner supportDEVELOPER CLAIM

Official partners WorkBuddy (including CodeBuddy) and OpenCode fully support V4.1-Flash.

WHAT CHANGED

Stored passport versions, without reconstructed history.

2026-09-30Passport created8 facts

DeepSeek launches DeepSeek‑V4.1‑Flash with new causal encoder–decoder and lower costs

DeepSeek announced DeepSeek-V4.1-Flash on 2026-09-10 as the smallest model in its new architecture family with native visual understanding.

Causal Encoder–Decoder architecture with approximately 8B active parameters for input and 16B for output.

Native visual understanding (vision-capable, multimodal).

Compared with the previous generation, V4.1-Flash reduces KV cache needs via compression, significantly cutting cache-hit charges and agent costs.

Tests by multiple parties and internal evaluation place V4.1-Flash ahead of DeepSeek-V4-Pro on performance, cost, speed, and total runtime.

V4-Flash and V4-Flash-Vision-Exp are retired; requests to deepseek-v4-flash and deepseek-v4-flash-vision-exp temporarily route to V4.1-Flash for compatibility.

Starting at 04:00 UTC on 2026-09-14, all deepseek-v4-pro requests will route to V4.1-Flash at V4.1-Flash rates until V4.1-Pro launches.

Official partners WorkBuddy (including CodeBuddy) and OpenCode fully support V4.1-Flash.

Optional analytics