[docs] Welcome page thumbnail and reranker fix (#3660)
This commit is contained in:
Parth Sharma
@@ -94,7 +94,7 @@ config = {
|
||||
"model": "gpt-4.1-nano-2025-04-14"
|
||||
}
|
||||
},
|
||||
"rerank": {
|
||||
"reranker": {
|
||||
"provider": "zero_entropy",
|
||||
"config": {
|
||||
"model": "zerank-1",
|
||||
|
||||
@@ -3,6 +3,10 @@ title: LLM as Reranker
|
||||
description: 'Flexible reranking using LLMs'
|
||||
---
|
||||
|
||||
<Warning>
|
||||
**This page has been superseded.** Please see [LLM Reranker](/components/rerankers/models/llm_reranker) for the complete and up-to-date documentation on using LLMs for reranking.
|
||||
</Warning>
|
||||
|
||||
LLM-based reranker provides maximum flexibility by using any Large Language Model to score document relevance. This approach allows for custom prompts and domain-specific scoring logic.
|
||||
|
||||
## Supported LLM Providers
|
||||
@@ -35,7 +39,7 @@ config = {
|
||||
"model": "gpt-4o-mini"
|
||||
}
|
||||
},
|
||||
"rerank": {
|
||||
"reranker": {
|
||||
"provider": "llm",
|
||||
"config": {
|
||||
"model": "gpt-4o-mini",
|
||||
@@ -69,7 +73,7 @@ Score from 0.0 to 1.0 where:
|
||||
|
||||
Provide only a single numerical score between 0.0 and 1.0."""
|
||||
|
||||
config["rerank"]["config"]["scoring_prompt"] = custom_prompt
|
||||
config["reranker"]["config"]["scoring_prompt"] = custom_prompt
|
||||
```
|
||||
|
||||
## Usage Example
|
||||
@@ -85,7 +89,7 @@ os.environ["OPENAI_API_KEY"] = "your-api-key"
|
||||
config = {
|
||||
"vector_store": {"provider": "chroma"},
|
||||
"llm": {"provider": "openai", "config": {"model": "gpt-4o-mini"}},
|
||||
"rerank": {
|
||||
"reranker": {
|
||||
"provider": "llm",
|
||||
"config": {
|
||||
"model": "gpt-4o-mini",
|
||||
@@ -145,7 +149,7 @@ Consider:
|
||||
Score from 0.0 to 1.0. Provide only the numerical score."""
|
||||
|
||||
config = {
|
||||
"rerank": {
|
||||
"reranker": {
|
||||
"provider": "llm",
|
||||
"config": {
|
||||
"model": "gpt-4o-mini",
|
||||
@@ -164,7 +168,7 @@ Use different LLM providers for reranking:
|
||||
```python Python
|
||||
# Using Anthropic Claude
|
||||
anthropic_config = {
|
||||
"rerank": {
|
||||
"reranker": {
|
||||
"provider": "llm",
|
||||
"config": {
|
||||
"model": "claude-3-haiku-20240307",
|
||||
@@ -176,8 +180,8 @@ anthropic_config = {
|
||||
|
||||
# Using local Ollama model
|
||||
ollama_config = {
|
||||
"rerank": {
|
||||
"provider": "llm",
|
||||
"reranker": {
|
||||
"provider": "llm",
|
||||
"config": {
|
||||
"model": "llama2:7b",
|
||||
"provider": "ollama",
|
||||
|
||||
@@ -1,68 +1,78 @@
|
||||
---
|
||||
title: Overview
|
||||
description: 'Pick the right reranker path to boost Mem0 search relevance.'
|
||||
---
|
||||
|
||||
Mem0 includes built-in support for various reranking providers to improve the relevance of memory search results. Rerankers post-process initial vector search results by re-scoring and re-ordering them using more sophisticated relevance models.
|
||||
Mem0 rerankers rescore vector search hits so your agents surface the most relevant memories. Use this hub to decide when reranking helps, configure a provider, and fine-tune performance.
|
||||
|
||||
## Usage
|
||||
<Info>
|
||||
Reranking trades extra latency for better precision. Start once you have baseline search working and measure before/after relevance.
|
||||
</Info>
|
||||
|
||||
To use a reranker:
|
||||
|
||||
1. **Configure**: Add a `rerank` configuration section in your memory config
|
||||
2. **Search**: Reranking is automatically enabled for all searches (default: `rerank=True`)
|
||||
|
||||
If no reranker is configured, search results will rely on vector similarity scoring alone.
|
||||
|
||||
For comprehensive configuration parameters for each reranker, please refer to [Config](./config).
|
||||
|
||||
### Controlling Reranking Per Search
|
||||
|
||||
Once configured, reranking is enabled by default. You can control it per-search:
|
||||
|
||||
```python
|
||||
# Reranking enabled (default)
|
||||
results = memory.search("query", user_id="user1")
|
||||
|
||||
# Explicitly enable reranking
|
||||
results = memory.search("query", user_id="user1", rerank=True)
|
||||
|
||||
# Disable reranking for this specific search
|
||||
results = memory.search("query", user_id="user1", rerank=False)
|
||||
```
|
||||
|
||||
## How Reranking Works
|
||||
|
||||
1. **Initial Search**: Vector similarity search retrieves candidate memories
|
||||
2. **Reranking** (if enabled): Selected reranker re-scores candidates using advanced models
|
||||
3. **Final Results**: Re-ordered results with both vector and rerank scores
|
||||
|
||||
<Note>
|
||||
Reranking operates as a post-processing step and can significantly improve search relevance at the cost of additional latency and API calls.
|
||||
</Note>
|
||||
|
||||
## Supported Rerankers
|
||||
|
||||
See the list of supported rerankers below.
|
||||
|
||||
<CardGroup cols={2}>
|
||||
<Card title="Zero Entropy" href="/components/rerankers/models/zero_entropy" />
|
||||
<Card title="Cohere" href="/components/rerankers/models/cohere" />
|
||||
<Card title="Sentence Transformer" href="/components/rerankers/models/sentence_transformer" />
|
||||
<Card title="Hugging Face" href="/components/rerankers/models/huggingface" />
|
||||
<Card title="LLM-based" href="/components/rerankers/models/llm" />
|
||||
<Card title="LLM Reranker" href="/components/rerankers/models/llm_reranker" />
|
||||
<CardGroup cols={3}>
|
||||
<Card
|
||||
title="Understand Reranking"
|
||||
description="See how reranker-enhanced search changes your retrieval flow."
|
||||
icon="search"
|
||||
href="/open-source/features/reranker-search"
|
||||
/>
|
||||
<Card
|
||||
title="Configure Providers"
|
||||
description="Add reranker blocks to your memory configuration."
|
||||
icon="settings"
|
||||
href="/components/rerankers/config"
|
||||
/>
|
||||
<Card
|
||||
title="Optimize Performance"
|
||||
description="Balance relevance, latency, and cost with tuning tactics."
|
||||
icon="speedometer"
|
||||
href="/components/rerankers/optimization"
|
||||
/>
|
||||
<Card
|
||||
title="Custom Prompts"
|
||||
description="Shape LLM-based reranking with tailored instructions."
|
||||
icon="code"
|
||||
href="/components/rerankers/custom-prompts"
|
||||
/>
|
||||
<Card
|
||||
title="Zero Entropy Guide"
|
||||
description="Adopt the managed neural reranker for production workloads."
|
||||
icon="sparkles"
|
||||
href="/components/rerankers/models/zero_entropy"
|
||||
/>
|
||||
<Card
|
||||
title="Sentence Transformers"
|
||||
description="Keep reranking on-device with cross-encoder models."
|
||||
icon="cpu"
|
||||
href="/components/rerankers/models/sentence_transformer"
|
||||
/>
|
||||
</CardGroup>
|
||||
|
||||
## When to Use Reranking
|
||||
## Picking the Right Reranker
|
||||
|
||||
- **Improved Relevance**: When vector search alone doesn't provide sufficiently relevant results
|
||||
- **Domain-Specific Queries**: For specialized terminology or context that benefits from advanced models
|
||||
- **Quality vs Speed Trade-off**: When you can accept higher latency for better search quality
|
||||
- **Production Systems**: Where search quality directly impacts user experience
|
||||
- **API-first** when you need top quality and can absorb request costs (Cohere, Zero Entropy).
|
||||
- **Self-hosted** for privacy-sensitive deployments that must stay on your hardware (Sentence Transformer, Hugging Face).
|
||||
- **LLM-driven** when you need bespoke scoring logic or complex prompts.
|
||||
- **Hybrid** by enabling reranking only on premium journeys to control spend.
|
||||
|
||||
Choose the reranker that best fits your use case:
|
||||
- **Zero Entropy**: Best balance of speed and quality for general use
|
||||
- **Cohere**: Enterprise-grade with excellent multilingual support
|
||||
- **Sentence Transformer**: Local deployment for privacy-sensitive applications
|
||||
- **Hugging Face**: Wide variety of pre-trained models for specialized use cases
|
||||
- **LLM-based**: Maximum customization with custom prompts and logic
|
||||
## Implementation Checklist
|
||||
|
||||
1. Confirm baseline search KPIs so you can measure uplift.
|
||||
2. Select a provider and add the `reranker` block to your config.
|
||||
3. Test latency impact with production-like query batches.
|
||||
4. Decide whether to enable reranking globally or per-search via the `rerank` flag.
|
||||
|
||||
<CardGroup cols={2}>
|
||||
<Card
|
||||
title="Set Up Reranking"
|
||||
description="Walk through the configuration fields and defaults."
|
||||
icon="settings"
|
||||
href="/components/rerankers/config"
|
||||
/>
|
||||
<Card
|
||||
title="Example: Reranker Search"
|
||||
description="Follow the feature guide to see reranking in action."
|
||||
icon="rocket"
|
||||
href="/open-source/features/reranker-search"
|
||||
/>
|
||||
</CardGroup>
|
||||
|
||||
Reference in New Issue
Block a user