Console

DeepSeek V4 Pro 0813

Context
1.05M
Max output
524K
Input → output
text → text
Weights
Open
Providers
7
01

Providers

7 providers · list price, latency and uptime
ProviderInput /1MCached /1MOutput /1MTTFB p50Uptime 24hRetention
1
DeepInfra
$1.30$0.10$2.60– – none
Context
1.05M
Max output
524K
Precision
fp8
Regions
default
Headquarters
–
Data retention
Zero retention
2
Nebius
$1.32$0.04$3.96– – none
Context
979K
Max output
524K
Precision
fp8
Regions
default
Headquarters
–
Data retention
Zero retention
3
Novita AI
$1.32$0.044$3.96– – none
Context
1.05M
Max output
393K
Precision
fp8
Regions
default
Headquarters
–
Data retention
Zero retention
Supported parameters
response_format
4
Sail Research
$0.92$0.04$2.77– – none
Context
1.05M
Max output
384K
Precision
fp4
Regions
default
Headquarters
–
Data retention
Zero retention
5
StreamLake
$1.32$0.044$3.96– – may retain
Context
1.05M
Max output
393K
Precision
–
Regions
default
Headquarters
–
Data retention
May retain prompts
6
Doubleword
$1.32$0.13$3.96– – may retain
Context
889K
Max output
–
Cache write /1M
$1.65
Precision
–
Regions
default
Headquarters
–
Data retention
May retain prompts
7
Boundless
$1.04$0.035$2.08– – none
Context
228K
Max output
228K
Precision
–
Regions
default
Headquarters
–
Data retention
Zero retention
Supported parameters
response_format

By default the router tries our listings in routing order and fails over before the first byte — never mid-answer. Pin, exclude or order providers with a routing policy. Each provider charges its own price: name one with provider.only and you pay its price; otherwise you pay the price of the provider that serves the request ($1.30 in / $2.60 out per 1M where routing starts, unless it fails over). provider.sort: "price" tries the provider with the lowest input + output price first, and provider.max_price leaves out a provider above your limit. Uptime is our own 24h measurement; a dash means too few requests to measure yet. Open a row for its specs, parameters and policies.

Boundless: Boundless serves this model at its own price, as any other provider does: a request that names no provider can be sent to it, and a request that names it with provider.only or provider.order chooses it.

Boundless's capacity is smaller than other providers'. A long prompt, or one without max_tokens, can be refused with a 413 that waiting will not fix: shorten the prompt, set max_tokens, or allow another provider.

Your prompt passes through the RouterPlus gateway before it reaches Boundless. Boundless is itself a router: it sends each call to one of its own backends, which RouterPlus neither chooses nor sees. Zero data retention is Boundless's commitment, not something RouterPlus measures. Boundless keeps a record of each request for 90 days, with account identifiers and no prompt or answer content.

02

Quickstart

any SDK — only the base URL changes
curl -N https://api.routerplus.com/v1/chat/completions \
  -H "Authorization: Bearer $TM_API_KEY" -H "content-type: application/json" \
  -d '{"model":"deepseek/deepseek-v4-pro-0813","stream":true,"max_tokens":256,
       "messages":[{"role":"user","content":"hi"}]}'
03

Pricing

per 1M tokens
prompt$1.3
cached prompt$0.1
completion$2.6

Other providers charge from $0.92 / $2.77 in / out per 1M: name one with provider.only to pay its price.

Every billed response carries usage.cost — recompute your bill from the wire.