Orca Models by Microsoft Research
About
The Orca family of large language models (LLMs), developed by Microsoft, emphasizes enhancing reasoning capabilities in smaller models. The original Orca model showcased impressive reasoning by emulating the step-by-step processes of more advanced LLMs 2. Orca 2, the latest version, builds upon this foundation, outperforming similar-sized models and rivaling the performance of much larger models on complex reasoning tasks 23. Both models are available for research, with Orca 2 particularly noted for its diverse reasoning techniques and adaptability to different tasks, demonstrating the potential of strategic training to enhance the capabilities of smaller LLMs 2.
Current Variants
Use-when guidance is derived from seed capabilities, context, release, and replacement fields.
Use when the workload needs 2k context and 13B parameters.
| Model | Use when | Released | Signals | Status |
|---|---|---|---|---|
| Orca 13B | Use when the workload needs 2k context and 13B parameters. | 2023-06 | 2k context13B parameters | Current |
Release Timeline
1 release groupSpecifications(1 models)
| Model | Released | Context | Parameters |
|---|---|---|---|
| Orca 13B | 2023-06 | 2k | 13B |
Frequently Asked Questions
- What is Orca used for?
- Orca is used for math-heavy prompts. The family description and listed model capabilities point to those workloads as the best fit.
- How does Orca compare to Harrier?
- Orca by Microsoft Research is strongest where you need math-heavy prompts, while Harrier by Microsoft Research is the closest related family to check for embedding. Orca has 1 listed variant and reaches up to 2k context, while Harrier reaches up to 33k context, so compare the specs and pricing tables before choosing a production model.
- Which Orca model should I use?
- If price is the main constraint, use the pricing table first because Orca does not have complete provider pricing in the local data. For the most capable/latest local choice, evaluate Orca 13B with 2k context.
