Entity Enricher offers two enrichment strategies that control how LLM calls are orchestrated. Choosing the right strategy affects accuracy, speed, and cost.
From the simplest to the most powerful configuration. Each builds on the previous one.
One model, one call. The entire schema is sent in a single prompt. Fast and cheap — ideal for simple schemas or quick iteration.
Aspirin
Any entity: company, drug, legal case, research paper...
Any entity type: company, drug compound, legal case, research paper, real estate property...
Use any LLM provider with your own API key.
Use any LLM provider. Bring your own API key, pay only for what you use.
Full schema in one call — auto-retries on validation failure.
If the LLM response doesn't match the schema, validation errors are sent back for automatic retry — up to 3 attempts.
Aspirin
Same strategy, but run across multiple models in parallel. Results are compared and arbitrated field-by-field to produce a single high-confidence output.
Aspirin
Any entity: company, drug, legal case, research paper...
Any entity type: company, drug compound, legal case, research paper, real estate property...
Catches type mismatches and ambiguities before wasting LLM credits on enrichment.
Match — Pharmaceutical Compound
Catches type mismatches before wasting LLM credits.
Bring your own API keys — works with any LLM provider.
Works with any LLM provider — Anthropic, OpenAI, Google, Azure, Ollama, and more. Use your own API keys, pay only for what you use.
Full schema in one call — auto-retries on validation failure.
If the LLM response doesn't match the schema, validation errors are sent back for automatic retry — up to 3 attempts.
An LLM examines each divergent field across models, selects the best value with reasoning, and produces the final trusted result.
Aspirin
ArbitratedReasoned field-level conflict resolution produces the final trusted result.
The schema is split by expertise domain. Each model receives focused sub-prompts for each domain. Results are deep-merged per model, then arbitrated across models. Maximum accuracy for complex, multi-domain schemas.
Aspirin
Any entity: company, drug, legal case, research paper...
Any entity type: company, drug compound, legal case, research paper, real estate property...
Catches type mismatches and ambiguities before wasting LLM credits on enrichment.
Match — Pharmaceutical Compound
Catches type mismatches before wasting LLM credits.
Bring your own API keys — works with any LLM provider.
Works with any LLM provider — Anthropic, OpenAI, Google, Azure, Ollama, and more. Use your own API keys, pay only for what you use.
Schema is split by domain. Each prompt is focused on one expertise area, producing more precise and deeper results.
Schema split by domain — self-correcting prompts retry on validation failure.
If the LLM response doesn't match the schema, validation errors are sent back for automatic retry — up to 3 attempts per prompt.
Deep merge of expertise responses per model.
Each model's expertise responses are merged into a single comprehensive result per model.
An LLM examines each divergent field across models, selects the best value with reasoning, and produces the final trusted result.
Aspirin
ArbitratedReasoned field-level conflict resolution produces the final trusted result.
| Aspect | Single Pass | Multi-Expertise |
|---|---|---|
| LLM Calls | 1 per model | N per model (1 per expertise domain) |
| Schema Sent | Full schema in one prompt | Subset per expertise domain |
| Execution | Sequential (one call) | Parallel (all expertises run simultaneously) |
| Accuracy | Good for simple schemas | Higher — focused prompts yield better results |
| Speed | Faster for small schemas | Parallel execution can be faster for large schemas |
| Cost | Lower (single call overhead) | Higher (multiple calls with per-call overhead) |
| Streaming | All-or-nothing result | Progressive — results stream as each expertise completes |
| Partial Success | No — entire call succeeds or fails | Yes — successful expertises are preserved, failed ones can be retried |
The multi-expertise strategy follows a four-step process for each model:
The schema is traversed recursively. Each property with an expertise domain tag is grouped with others sharing the same domain. For example, revenue and market_cap go to the “financial” group, while gmp_status goes to “regulatory”.
Each expertise group becomes a minimal sub-schema containing only its properties. This means the LLM receives a smaller, more focused prompt and only needs to fill in fields it specializes in.
All expertise calls run concurrently. A schema with 5 expertise domains will launch 5 LLM calls at the same time. As each one completes, its results are deep-merged into the accumulated output and streamed to the UI in real-time.
If some expertises fail, the system returns the merged output from successful ones with a “Partial” status. You can retry only the failed expertises, and the new results will be merged into the existing output without repeating the work that already succeeded.
Both strategies can be combined with multi-model enrichment. When you select multiple models, each model runs the chosen strategy independently. The results can then be merged using multi-model fusion to produce a single high-confidence output.