/ Models

Language models

DeepSeek: DeepSeek V4 Flash 0731

DeepSeek V4 Flash 0731 is a sparse mixture-of-experts model from DeepSeek, with 13B active parameters out of 284B total. This re-post-trained revision is suited for coding, reasoning, and agent workflows.

Maker
Deepseek
Context window
1M
Max output
384K
Input / 1M
$0.08
Output / 1M
$0.18
1M in + 1M out
$0.26

Everything else

Max output384K
Input / 1M$0.08
Output / 1M$0.18
Cache read / 1M$0.016
Cache write / 1M
1M in + 1M out$0.26
Takestext
Returnstext
Toolsyes
Reasoningoptional · max, high, low
Open weightsyes
Released2026-07-31
Knowledge cutoff
Retires
Catalog iddeepseek/deepseek-v4-flash-0731

Benchmarks

sourceepoch
eci152.53
confidenceLikely
accessibilityOpen weights (unrestricted)
gpqa0.91
gpqaStderr0.0177
aime0.944
aimeStderr0.0209
simpleQA0.347
simpleQAStderr0.0151

Benchmark scores by Epoch AI, CC BY 4.0.

Using it

In 00, this model is picked per agent — and per tier, so the model that answers your customers need not be the one that writes your code. Get 00.

← all models