Compositev12
← All models Model directory

AI model

DeepSeek: DeepSeek V4 Flash 0731

deepseek/deepseek-v4-flash-0731

DeepSeek: DeepSeek V4 Flash 0731 is a 1,048,576-token model served on Composite at $0.0750 per million tokens in and $0.225 per million tokens out. Ranked #15 of 196 models by traffic on Composite over the last 30 days, taking 1.93% of tokens and 2.78% of requests (this model and its routing variants). Measured on Composite, 100.0% of its last 2 requests on Composite succeeded, the typical first token arrives in 17575 ms, output runs at about 90.4 tokens per second.

Try in Playground

Community rating

— No community ratings yet.

Sign in to rate this model. Ratings come from accounts that have run requests through Composite.

Specifications

Specifications and pricing as served by Composite
Model IDdeepseek/deepseek-v4-flash-0731
Providercv11
Context window1,048,576 tokens
Acceptstext
Returnstext
Input price$0.0750 per million tokens
Output price$0.225 per million tokens
Prompt cachingSupported — cached input is billed at the provider's cache rate, on credits and plans alike
ThinkingOptional - see "DeepSeek: DeepSeek V4 Flash 0731 (Reasoning)" in the catalog to turn it on by default
ModeratedNo
Cost on a monthly planAbout 0.7 plan credits per typical call (8k in, 1k out)
Routing endpoints10 (base plus 9 routing variants)

Traffic rank on Composite

#15 of 196

1.93% of tokens and 2.78% of requests over 30 days, including routing variants.

Success rate

100.0%

Across the last 2 requests Composite served for this model.

Time to first token

17575 ms

Median across the same sample.

Output speed

90.4 tok/s

Median generation rate after the first token.

Routing variants

The same model behind 9 alternative endpoints. Call the id directly to pin a route; the base id above picks for you.

RouteModel IDProviderPrice
Normal deepseek/deepseek-v4-flash-0731-normal cv11 $0.0200 per million tokens input · $0.800 per million tokens output
Normal + cheapThink deepseek/deepseek-v4-flash-0731-normal:cheapThink cv11 $0.0200 per million tokens input · $0.800 per million tokens output
Cheapest provider deepseek/deepseek-v4-flash-0731:cheapest-provider DeepInfra $0.0750 per million tokens input · $0.225 per million tokens output
Cheapest provider + cheapThink deepseek/deepseek-v4-flash-0731:cheapest-provider:cheapThink DeepInfra $0.0750 per million tokens input · $0.225 per million tokens output
cheapThink deepseek/deepseek-v4-flash-0731:cheapThink cv11 $0.0750 per million tokens input · $0.225 per million tokens output
Fast deepseek/deepseek-v4-flash-0731:fast DeepInfra $0.0750 per million tokens input · $0.225 per million tokens output
Fast + cheapThink deepseek/deepseek-v4-flash-0731:fast:cheapThink DeepInfra $0.0750 per million tokens input · $0.225 per million tokens output
Quality deepseek/deepseek-v4-flash-0731:quality Parasail $0.175 per million tokens input · $0.350 per million tokens output
Quality + cheapThink deepseek/deepseek-v4-flash-0731:quality:cheapThink Parasail $0.175 per million tokens input · $0.350 per million tokens output

Pricing

$0.0750 per million tokens input · $0.225 per million tokens output

Credits pay the provider’s price plus 25%; cache hits cost less.

On a monthly plan, a typical call (8k tokens in, 1k out) uses about 0.7 plan credits, at the provider's own price.

Provider

cv11

Roleplay fit

DeepSeek models are known for strong reasoning and coding ability at aggressive pricing, which makes them a common pick for roleplay accounts that send a high volume of messages.

Context behavior

With a 1,048,576-token context window, it can hold a long-running roleplay or a full lorebook in memory without losing earlier plot details, so it suits multi-session campaigns and character cards with heavy backstory.

Cost for regular use

It is priced low enough for high-volume daily roleplay without the per-message cost adding up quickly.

From the provider

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....

Supplied by cv11, not written by Composite.