wehub-resource-sync
39 lines
877 B
TOML
39 lines
877 B
TOML
name = "DeepSeek V4 Flash"
|
|
description = "Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work"
|
|
# OpenAI: `thinking.type = enabled|disabled`, `reasoning_effort = high|max`.
|
|
# Anthropic: `thinking.type`, `output_config.effort = high|max`; budget ignored.
|
|
# https://api-docs.deepseek.com/api/create-chat-completion (accessed 2026-06-25)
|
|
family = "deepseek-flash"
|
|
release_date = "2026-04-24"
|
|
last_updated = "2026-04-24"
|
|
attachment = false
|
|
reasoning = true
|
|
temperature = true
|
|
knowledge = "2025-05"
|
|
tool_call = true
|
|
structured_output = true
|
|
open_weights = true
|
|
|
|
[[reasoning_options]]
|
|
type = "toggle"
|
|
|
|
[[reasoning_options]]
|
|
type = "effort"
|
|
values = ["high", "max"]
|
|
|
|
[interleaved]
|
|
field = "reasoning_content"
|
|
|
|
[cost]
|
|
input = 0.14
|
|
output = 0.28
|
|
cache_read = 0.0028
|
|
|
|
[limit]
|
|
context = 1_000_000
|
|
output = 384_000
|
|
|
|
[modalities]
|
|
input = ["text"]
|
|
output = ["text"]
|