impulse

Search

Open weightsSearch
API id search
Input / 1M
-
not listed
Output / 1M
-
not listed
Context window
-
tokens
Max output
-
tokens per reply
Hosts
19
19 listings
Released
-
date not published

Where to run it

19 hosts serve Search. The comparison is on blended price, 3 parts input to 1 part output.

HostIn / out per 1M
DeepSeekLab API
search
-
search
-
search
-
search
-
search
-
search
-
Exa AICloud
search
-
search
-
search
-
LinkupCloud
search
-
NimbleCloud
search
-
search
-
SearXNGCloud
search
-
SerperCloud
search
-
TavilyCloud
search
-
search
-
search
-
You.comCloud
search
-
PerplexityGateway
search
-

Capabilities

Core

  • Streaming
  • Function calling
  • JSON mode
  • Structured outputs

Multimodal

  • Vision
  • Image input
  • Audio
  • Image generation
  • Video

Advanced

  • Reasoning
  • Tool calling
  • MCP
  • Prompt caching
  • Batch API
  • Embeddings
Takes in
Text
Produces
Text
Open weights
License: DeepSeek License (MIT code)
Runtimes: SGLang, TGI, vLLM

Similar models

This model has no listed price, so these are current models with similar capabilities.

Smaller DeepSeek models

No similar DeepSeek model keeps its key capabilities.

Other frontier labs

No similar match from another frontier lab.

Open-weight picks

No similar open-weight match.

Call it

Using DeepSeek's API.

Python
from litellm import completion

response = completion(
    model="deepseek/search",
    messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)

Quick answers

Which APIs offer Search?

Search runs on 19 APIs: DeepSeek, AgentCore, APISerpent, Bing Grounding, DataForSEO, DuckDuckGo, Exa AI, Firecrawl, Google PSE and 10 more.

Is Search being retired?

No retirement date is scheduled for it on any API we track.

What is Search's API model id?

Use search with DeepSeek.

Sources

Raw catalog entry for search on DeepSeek
{
  "input_cost_per_query": 0.00075,
  "litellm_provider": "deepseek",
  "mode": "search"
}