+
+
DeepSeek V4.1 Flash 1017
deepseek/deepseek-v4-1-flash-1017
Share
DeepSeek V4.1 Flash 1017 delivers high-speed multi-head latent reasoning, providing ultra-fast code and math problem solving at fraction-of-cent token economics.
+
+
Pricing
USD, pay per token$0.15
per 1M input tokens$0.01
per 1M cached input tokens$0.60
per 1M output tokens+
+
Measured performance
Output speed
320.5
output tok/s, model onlyTime to first token
410
milliseconds to first reasoning tokenCache hit rate
29.7%last 24 hours
+
+
Access
Confirm how your client reaches this model.
Provider
DeepSeek
API compatibility
Mission-compatible Outcome Intents
Anthropic compatibility
Anthropic-compatible Messages, POST /v1/messages
Accepted input
Text, code, and structured data
Availability
Available
Data retention
Your state and execution memory are never stored or used for external model training.
Capacity
Check the limits your workload must fit.
Context window
128,000 tokens
Maximum request size
2.5 MB per request
Maximum generated output
16,384 tokens
Capabilities
See which request modes the API supports.
Tool calling
Supported
JSON mode
Supported
Streaming
Supported
Precision
FP8, vendor-native release
View Full Spec
+
+
Code examples
Set HOROS_GATEWAY_KEY to your workspace API key before using an example.
CLIENT_REQUEST_ID="$(uuidgen)"
curl -X POST "https://api.horos.com/v1/chat/completions" \
-H "Authorization: Bearer $HOROS_GATEWAY_KEY" \
-H "X-Client-Request-Id: $CLIENT_REQUEST_ID" \
-H "Content-Type: application/json" \
-d '{"model":"deepseek-v4-1-flash-1017","messages":[{"role":"user","content":"Hello"}],"max_tokens":2048}'+
+
Trust and provenance
Verify the company and operating credentials behind this capability.
HOROS is a product of System Integration.