ZappushZappush
SkillsUse CasesBenchmarkCommunity
← DeepSeek Models

DeepSeek

Deepseek V4 Flash

deepseek/deepseek-v4-flash

1.0M context·65.5K max output·text->text

Input

$0.0900

per 1M tokens

Output

$0.1800

per 1M tokens

Cache Read

$0.002800

per 1M tokens

Cache Write

$0.00

per 1M tokens

Quality Benchmarks

FrontierValue: -1505.8

Coding

Code generation, completion, debugging

57.7

Math

Competition math, calculus, olympiad

79.6

Reasoning

Logic, spatial, puzzle solving

71.7

Language

Summarization, paraphrase, writing

66.6

Overall

68.9

Source: LiveBench (livebench.ai) / Scores 0-100

Capabilities

VisionTool UseReasoningPrompt Caching

Similar Price Range

Qwen3 Next 80b A3b Instruct

Qwen

$0.0900 in$1.10 out262.1K

Qwen3 235b A22b 2507

Qwen

$0.0900 in$0.1000 out262.1K

Qwen3 Next 80b A3b Thinking

Qwen

$0.0975 in$0.7800 out262.1K

More from DeepSeek

Deepseek Chat V3 0324

$0.2000 / $0.7700

Deepseek Chat

$0.2002 / $0.8001

Deepseek Chat V3.1

$0.2100 / $0.7900

Deepseek V3.2

$0.2288 / $0.3432

Deploy your AI agent for $33/mo flat.

Managed Telegram bot hosting. We handle the infrastructure.

Any model. Any channel. Zero infrastructure.

Deploy Your Agent
Zappush© 2026 Zappush
HomeSkillsUse CasesBenchmarkGuarantee