How EmaFusion™ Works
EmaFusion™ is Ema's proprietary system of models. It routes across 40+ large language models from 9+ providers (including OpenAI GPT and o-series, Anthropic Claude, Google Gemini, Meta Llama, Mistral, Moonshot Kimi, Alibaba Qwen, DeepSeek, and enterprise-trained custom models) to deliver optimal results for every sub-task an AI Employee performs.
Architecture
When a task is created, the EmaFusion™ model recognizes the sub-task type and selects the single best model for that task's framework from its broad catalog of providers. Unlike AI tools locked to one model, EmaFusion™ dynamically picks the right model for each sub-task. If that model's response falls below a confidence threshold, it cascades to the next-best candidate. Cross-model validation across multiple models only runs in select high-stakes scenarios; routing to a single best model is the default behavior.
Routing logic
- Task analysis: EmaFusion™ examines the incoming sub-task (e.g., conversation generation, structured query, reasoning, summarization).
- Model selection: Based on benchmarked performance, cost, and latency data, EmaFusion™ ranks candidate models and picks the single best one for the sub-task.
- Execution with cascading fallback: The selected model processes the task. If its response falls below the confidence threshold for the chosen optimization mode, EmaFusion™ cascades to the next-best model in the ranked list.
- Continuous improvement: Ema regularly benchmarks and updates model performance data, so routing decisions improve over time.
Supported model families
EmaFusion™ routes across a broad set of model families. The following are the primary families available in the current release:
| Provider | Models | Typical strengths |
|---|---|---|
| OpenAI (GPT) | GPT-5, GPT-5-mini, GPT-5.2, GPT-5.4, GPT-5.4-mini, GPT-5.6-Luna, GPT-5.6-Sol, GPT-5.6-Terra, GPT-4.1, GPT-4.1-mini, GPT-4o, GPT-4o-mini, and earlier variants | General-purpose generation, structured output, tool calling |
| OpenAI (o-series) | o1, o3, o3-mini, o4-mini | Advanced reasoning, step-by-step problem solving, math and code |
| Anthropic | Claude Opus (4.1 through 5), Claude Sonnet (4.5, 4.6, and Sonnet 5), Claude Haiku (3.x and 4.5) | Reasoning, long-context tasks, safety-sensitive content |
| Gemini Pro and Gemini Flash (2.5, 3, 3.1, and 3.5 variants, including Flash-Lite) | Structured queries, multimodal tasks, translation | |
| Meta | Llama 3.1, Llama 3.3 | Cost-effective generation, open-weight flexibility |
| Mistral | Mistral Large | Multilingual tasks, code generation |
| Moonshot | Kimi K2.6, Kimi K2.7, Kimi K3, Kimi K3 Fast | Long-context understanding, agentic tasks |
| Z.ai | GLM 5.1, GLM 5.2 | Reasoning, agentic tasks, cost efficiency |
| DeepSeek | DeepSeek V4 Pro, DeepSeek V4 Flash | Reasoning, code generation, cost efficiency |
| Alibaba | Qwen | Multilingual generation, code |
| Sarvam | Sarvam-M, Sarvam 30B, Sarvam 105B | Indic-language tasks, India data residency |
| Custom (BYOM) | Your private models | Domain-specific, compliance-restricted tasks |
Benefits
| Benefit | Description |
|---|---|
| Maximize accuracy, minimize cost | Uses the right model for each sub-task, avoiding the cost of always defaulting to a single premium model for every request. |
| Continuous improvement | Ema benchmarks and updates model performance regularly. New models are integrated as they become available. |
| Future-proof | New models are added with no vendor lock-in. Your AI Employees automatically benefit from advances in the LLM ecosystem. |
| Fewer hallucinations | Outputs can be cross-checked across multiple models, reducing the likelihood of hallucinated content. |
| Privacy by design | Ema automatically obfuscates sensitive data (names, emails, phone numbers) before sending to models. |
Bring your own model (BYOM)
Instead of routing through the EmaFusion™ catalog, an AI Employee can run only on the models from your own configured LLM providers. This is useful for:
- Keeping all traffic on a specific deployment, such as your own Azure OpenAI resource.
- Models accessed through your own accounts for compliance or cost reasons.
BYOM is an AI Employee-wide choice: it applies to every agent under the AI Employee, and individual agents cannot override it. See EmaFusion™ setup for instructions on configuring BYOM and providers.
Example use cases
- Customer Support AI Employee: Uses GPT-5.4 for conversation, Claude for reasoning, and Gemini for structured queries.
- Finance Analyst AI Employee: Routes to an enterprise-trained proprietary model for compliance-sensitive tasks, with fallback to GPT-4o for summaries.
- Healthcare AI Employee: Uses a private BYOM model with a HIPAA-compliant API for patient data.
- Voice AI Employee: Routes speech-to-text diarization to GPT-5.4 for high-accuracy transcription, with downstream agents using EmaFusion™'s balanced routing for response generation.
Configuration hierarchy
EmaFusion™ configuration is resolved through a two-level hierarchy: a default set at the AI Employee level, and optional per-agent overrides that replace the default for a specific agent in a workflow.
- AI Employee default. Every AI Employee has a default EmaFusion™ configuration (the selected models, optimization mode, and any BYOM settings) that all agents use unless they override it.
- Agent-level inheritance and override. Agents inherit the AI Employee default and can unbind from it with the Override default configuration toggle to set an agent-specific model or optimization mode.
For step-by-step configuration of the AI Employee default, per-agent overrides, and providers, see EmaFusion™ setup.
References
What's next
- EmaFusion™ setup: choose a model-selection mode and optimization priority, override per agent, and configure providers.
- EmaFusion™ overview: what EmaFusion™ is and why it matters.