MODELS / DEEPSEEK

DeepSeek V4.1 Flash 1017 Ultrafast

DeepSeek V4.1 Flash 1017 Ultrafast serves the October 17, 2026 update of DeepSeek V4.1 Flash through Fireworks' accelerated serverless endpoint. It supports text and image input, configurable reasoning, tool calling, structured outputs, and automatic prefix caching. Available…

TEXTREASONINGIMPLICIT-CACHINGSTRUCTURED-OUTPUTTOOL-USEVISIONOPEN-WEIGHTS
1.0M
CONTEXT WINDOW
-
MAX OUTPUT TOKENS
$2.40
INPUT PRICE / 1M TOKENS
$9.60
OUTPUT PRICE / 1M TOKENS
QUICKSTART
PROTOCOL:
OPENAI COMPATIBLE
STREAMING:
SUPPORTED

Call it in one request.

OpenAI-compatible: change the base URL, and the setup is complete.

curl https://enterprise.blackbox.ai/v1/chat/completions \
  -H "Authorization: Bearer $BLACKBOX_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "blackboxai/deepseek/deepseek-v4.1-flash-1017-ultrafast",
    "messages": [{ "role": "user", "content": "Hello!" }],
    "stream": true
  }'
SPECIFICATIONS
CONTEXT:
1.0M
PROVIDER:
DEEPSEEK
TYPE:
TEXT

The full spec sheet.

Context, pricing, modalities, routing, and deployment: everything DeepSeek V4.1 Flash 1017 Ultrafast supports, in one table.

MODEL ID
blackboxai/deepseek/deepseek-v4.1-flash-1017-ultrafast
DEVELOPED BY
DeepSeek
MODEL TYPE
Text
RELEASED
Nov 6, 2026
CONTEXT WINDOW
1.0M
MAX OUTPUT TOKENS
-
INPUT MODALITIES
text, image
OUTPUT MODALITIES
text
INPUT PRICE
$2.40 / 1M tokens
OUTPUT PRICE
$9.60 / 1M tokens
CACHE READ
$0.048 / 1M tokens
REASONING
Supported
DATA RETENTION
Standard retention
DEDICATED DEPLOYMENT
Enterprise · single-tenant
SUPPORTED PARAMETERS
max_tokenstemperaturestoptoolstool_choicereasoninginclude_reasoningresponse_formatstructured_outputs