What is the IQ of deepseek-v4-flash-vision-exp?

deepseek-v4-flash-vision-exp has an estimated IQ of 122, which ranks #36 out of 164 public models on the AI IQ leaderboard. The score combines seven equally weighted reasoning dimensions using measured benchmark results and conservative estimates for missing coverage.

How much does deepseek-v4-flash-vision-exp cost?

deepseek-v4-flash-vision-exp costs $0.44 per million input tokens and $1.32 per million output tokens. Adjusted for how many tokens it actually consumes on real tasks, its effective cost works out to $2.84 per million input-plus-output tokens.

Is deepseek-v4-flash-vision-exp open source?

No — deepseek-v4-flash-vision-exp is a proprietary model. Its weights are not publicly available, so it is used through hosted APIs and products.

When was deepseek-v4-flash-vision-exp released?

deepseek-v4-flash-vision-exp was publicly released on August 21, 2026.

How large is the context window of deepseek-v4-flash-vision-exp?

deepseek-v4-flash-vision-exp supports a context window of 1,000,000 tokens — the amount of text it can consider in a single request.

How fast is deepseek-v4-flash-vision-exp?

In independent measurements, deepseek-v4-flash-vision-exp generates about 189 tokens per second, with a typical response time of about 14 seconds.

DeepSeek

deepseek-v4-flash-vision-exp

Public AI IQ model profile with source-backed benchmark results and derived ranking context.

IQ

122

EQ

n/a

Effective Cost

$2.8443

IQ Rank

#36

deepseek-v4-flash-vision-exp API pricing — full price table, workload cost examples, and cheaper alternatives.

Release and Capabilities

Public release	2026-08-21
Modalities	text, image
Context window	1,000,000 tokens

Experimental model available through the DeepSeek API as deepseek-v4-flash-vision-exp

Publisher-reported capabilities

The model accepts mixed text and image inputs through Chat Completions, Messages, and Responses APIs.
DeepSeek reports text capabilities on par with DeepSeek V4 Flash and a substantial multimodal-agent improvement over the text-only model.
The API supports a 1M-token context window and a maximum 384K combined thinking-and-output budget.

Limitations and availability caveats

DeepSeek labels the model experimental and has not announced open weights.
The launch benchmark values use DeepSeek's own harness at max effort and are not imported into independent sparse benchmark files.
FIM completion is not supported on the vision endpoint.

Official release and model references

Measured Coverage and Estimates

Deepseek V4 Flash Vision Exp has 12 of 39 tracked IQ benchmark inputs across 4 of 7 dimensions. Raw benchmark tables contain source-backed results only. Derived IQ combines all six dimensions using the shared scoring methodology, including estimates where direct coverage is incomplete. Configured predecessor references: Deepseek V4 Flash. A reference is an estimation input, not evidence that this model has the predecessor's measured results.

Among dimensions with direct benchmark coverage, this profile's highest estimated scores are Programmatic Reasoning (IQ 129) and Academic Reasoning (IQ 127). These dimension estimates can include statistical adjustments or predecessor priors; they are not raw benchmark percentages or claims of real-world task success.

Benchmark versions, evaluation harnesses, and coverage affect comparisons; a missing result is not a score of zero. See the scoring methodology for estimation and uncertainty limits.

IQ Dimensions

Dimension	IQ	Coverage
Abstract Reasoning	102	0/5
Mathematical Reasoning	131	0/5
Academic Reasoning	127	5/6
Programmatic Reasoning	129	1/6
Computer Use	123	0/5
Reliability	115	2/6
Knowledge Work	124	4/6

Benchmark Results

Benchmark	Score	Dimension
AA-Briefcase v1.1 Elo	1424.63	Knowledge Work
AA Long Context Reasoning v1.1	81.3333	Reliability
AA Omniscience	-17.6333	Reliability
AutomationBench-AA	47.4709	Knowledge Work
CritPt	10.8571	Academic Reasoning
GDP.pdf	12	Knowledge Work
GDPval-AA v2.1 normalized score	52.3355	Knowledge Work
GPQA Diamond	91.3131	Academic Reasoning
Humanity's Last Exam	34.4764	Academic Reasoning
MMMU-Pro	74.7977	Academic Reasoning
SciCode	49.6528	Academic Reasoning
Terminal-Bench 4.0	12.1212	Programmatic Reasoning

Compare

Frequently Asked Questions

What is the IQ of deepseek-v4-flash-vision-exp?: deepseek-v4-flash-vision-exp has an estimated IQ of 122, which ranks #36 out of 164 public models on the AI IQ leaderboard. The score combines seven equally weighted reasoning dimensions using measured benchmark results and conservative estimates for missing coverage.
How much does deepseek-v4-flash-vision-exp cost?: deepseek-v4-flash-vision-exp costs $0.44 per million input tokens and $1.32 per million output tokens. Adjusted for how many tokens it actually consumes on real tasks, its effective cost works out to $2.84 per million input-plus-output tokens.
Is deepseek-v4-flash-vision-exp open source?: No — deepseek-v4-flash-vision-exp is a proprietary model. Its weights are not publicly available, so it is used through hosted APIs and products.
When was deepseek-v4-flash-vision-exp released?: deepseek-v4-flash-vision-exp was publicly released on August 21, 2026.
How large is the context window of deepseek-v4-flash-vision-exp?: deepseek-v4-flash-vision-exp supports a context window of 1,000,000 tokens — the amount of text it can consider in a single request.
How fast is deepseek-v4-flash-vision-exp?: In independent measurements, deepseek-v4-flash-vision-exp generates about 189 tokens per second, with a typical response time of about 14 seconds.