Relace Search
VerifiedLong-context LLM built for precise search over massive text inputs.
About Relace Search
Relace Search is a proprietary large language model whose weights and parameter count are not publicly disclosed. Its defining technical feature is the 256000-token context window, which enables processing of extended text sequences without early truncation. This architecture is optimized for tasks that depend on maintaining context across lengthy inputs.
Strengths center on sustained coherence during search and retrieval operations involving large documents or corpora. Because the model is closed-weight and text-only, it delivers focused performance on semantic matching and information extraction without multimodal overhead.
Common applications include enterprise document search, long-form content analysis, and retrieval-augmented systems that must reference earlier sections of extensive texts. Integration typically occurs through hosted APIs where the large context can be leveraged for accurate, context-aware results.
Capabilities
How Relace Search compares
Relace Search (striped bar) vs other language models on intelligence, speed and price.
Price
USD per 1M output tokens · Lower is better · Relace Search ranks #29 of 34
Sources: Artificial Analysis (intelligence, speed) · OpenRouter (price).
Best for
Long-context document retrieval
Relace Search handles queries over extensive inputs thanks to its 256000-token context, enabling direct search across full reports or repositories without chunking.
Semantic search over knowledge bases
The model supports intent-aware lookup within large structured or unstructured collections while preserving surrounding context for accurate results.
Multi-turn research sessions
Users can maintain extended conversations that reference prior information spread across thousands of tokens for iterative exploration.
Strengths & limitations
Strengths
- +Handles very large text inputs
- +Search-oriented design
- +Pure text focus for efficiency
Limitations
- –Text modality only
- –No built-in vision or code execution
- –Standard LLM hallucination risks
Cost calculator
Estimate what Relace Search would cost for your usage.
Based on Relace Search's $1.00/1M input · $3.00/1M output. Estimate only — actual cost varies by provider and caching.
Quick start
OpenRouter's API is OpenAI-compatible — most SDKs work by just swapping the base URL. Only the model slug changes between models.
import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://openrouter.ai/api/v1",
apiKey: process.env.OPENROUTER_API_KEY,
});
const completion = await client.chat.completions.create({
model: "relace/relace-search",
messages: [{ role: "user", content: "Hello!" }],
});
console.log(completion.choices[0].message.content);Model slug: relace/relace-search
Editor's verdict
Relace Search is Relace's proprietary language models with a 256K-token context window.
At $3.00 per 1M output tokens, it is mid-priced for its class.
It is available through Relace's API and aggregators like OpenRouter.
Best suited to handles very large text inputs and search-oriented design.
Frequently asked questions
The model provides a context window of 256000 tokens.
User reviews
Real, verified reviews from the community shape this model's rating.
Loading reviews…
Other Relace models
Sibling versions in the Relace family from Relace.
Similar models
Other language models worth comparing.