DeepSeek: DeepSeek V4 Flash Vision Exp
LiveDeepSeek1M context · Text, Vision, Tools, ReasoningBenchmark price · $0.2156 input / $0.6468 output per 1M tokens
Best discount now60.0%offYou pay 40.0% of reference.
Median discount60.0%offAcross 1 live offer.
Realized · 7 days59.9%avgRequest-weighted average.
Requests · 24h1171 offer from 2 sellers.
Current offers
1 of 9 offers
Availability
Providers
ProviderDiscountPriceLatencyUptimeCacheSold
OpenRouter
60%
$0.0862 in
$0.26 out
679ms
100.0%
10.7%
-
Total
Cumulative offers
Sold
$0
Discount trend
Realized · 7 days61.9%
Best, twice dailyMedian 60.0%
Recent fills
10:38100 tokensOpenRouter60.7%
10:10587 tokensOpenRouter60.2%
09:55100 tokensOpenRouter100.0%
09:51494 tokensOpenRouter60.3%
09:33100 tokensOpenRouter60.7%
09:33100 tokensOpenRouter60.7%
09:01250 tokensOpenRouter60.2%
08:5547 tokensOpenRouter100.0%
08:2847 tokensOpenRouter58.8%
08:07588 tokensOpenRouter59.8%
07:55100 tokensOpenRouter100.0%
07:5427,944 tokensOpenRouter60.0%
07:5327,956 tokensOpenRouter60.0%
07:5327,500 tokensOpenRouter60.0%
07:23419 tokensOpenRouter59.9%
07:23773 tokensOpenRouter59.9%
07:23100 tokensOpenRouter60.7%
07:21247 tokensOpenRouter60.7%
06:591,222 tokensOpenRouter59.9%
06:59693 tokensOpenRouter59.6%
Times in UTCSettled on Base in USDC.
Performance by provider route · 24h
Measured on Surplus traffic| Route | p50 latency | p95 latency | Throughput | Success rate | ShareShare of traffic |
|---|---|---|---|---|---|
| OpenRouter | 1.24 s | 5.55 s | — | 100.0% | 100% |
OpenRouter100% of traffic
100.0% success
Use this modelChange your base URL.
Change your base URL.
Keep your code.
Any OpenAI-compatible client works. Add /min30 to the path to only buy at 30% off or better.
Read the quickstart ↗Parameters this model acceptsPOST /v1/chat/completions
- modelstringThis model’s id
- messagesarrayThe conversation, as role and content pairs
- streambooleanSend the answer as server-sent events
- max_tokensintegerMost output tokens to generate
- temperaturenumberSampling temperature
- top_pnumberNucleus sampling
- stopstring | arrayStop sequences
- toolsarrayFunctions the model may call, in the OpenAI format
- tool_choicestring | objectauto, required, none, or one named function
- response_formatobjectJSON mode or a JSON schema for structured output
- reasoning_effortstringHow hard to think, as a string such as "high"
- reasoningobjectReasoning options as an object, such as {"effort": "high"}
- include_reasoningbooleanStripped before the request reaches the provider
- frequency_penaltynumberPenalises tokens by how often they appeared
- presence_penaltynumberPenalises tokens that appeared at all
- logprobsbooleanReturn token log probabilities
- top_logprobsintegerHow many top tokens to return probabilities for
Context window 1M tokens · Max output 384K tokensOthers are passed through; most providers ignore what they don’t support. Parameter compatibility ↗
from openai import OpenAI client = OpenAI( api_key="inf_your_key_here", base_url="$SURPLUS_BASE_URL" ) response = client.chat.completions.create( model="deepseek-v4-flash-vision-exp", messages=[{"role": "user", "content": "Hello!"}] )
SURPLUS_BASE_URL
https://api.surplusintelligence.ai/v1