DeepSeek-V4-Flash-Vision-Exp (Reasoning)

by DeepSeek

DeepSeek-V4-Flash-Vision-Exp (Reasoning) is a reasoning model from DeepSeek, released on August 21, 2026. It has a 1M-token context window and can generate up to 1M output tokens. API pricing is $0.44 per 1M input tokens and $1.32 per 1M output tokens. It ranks #32 of 220 models by Intelligence Score with a score of 44.7.

DeepSeek-V4-Flash-Vision-Exp is an experimental multimodal vision-understanding model on the DeepSeek API (`model='deepseek-v4-flash-vision-exp'`). It accepts text and image inputs and produces text. DeepSeek reports pure-text agent, reasoning, and world-knowledge performance on par with official DeepSeek-V4-Flash, with a large jump on agent benchmarks that require visual understanding. Thinking mode is default and supports low / high / max effort; the model supports JSON output, tool calls, the Responses API, and the Anthropic API. FIM completion is not supported. Context length is 1M tokens with up to 384K max output. Same peak/off-peak token pricing as V4-Flash; images are converted to input tokens (upper bound 384 tokens per image after resize).

Key specs

Compare on the AI leaderboard · All AI models

More models from DeepSeek

• DeepSeek-V4.1-Flash (Reasoning) — DeepSeek
• DeepSeek-V4-Pro-0813 (Reasoning) — DeepSeek
• DeepSeek-V4-Flash-0423 (Reasoning) — DeepSeek
• DeepSeek-V3.2 (Thinking) (Reasoning) — DeepSeek
• DeepSeek-R1-0528 — DeepSeek
• DeepSeek R1 Zero — DeepSeek