Palmyra Models by Writer
About
The Palmyra family of large language models (LLMs), developed by Writer, consists of several models tailored for diverse tasks, including Palmyra-Large, a 20-billion parameter causal decoder-only model trained on a massive dataset of over 800 billion tokens. This foundational model adeptly handles nuanced tasks such as sentiment classification and summarization. Writer has also introduced specialized models like Palmyra-Med for healthcare and Palmyra-Fin for finance, which achieve leading benchmark results in their respective areas. These models are ideally suited for precision in regulated industries and are available via Writer's platform, APIs, and open-model licenses. The Palmyra models stand out for their efficiency, utilizing cost-effective training techniques to maintain high accuracy even with relatively smaller parameter counts.
Current Variants
Use-when guidance is derived from seed capabilities, context, release, and replacement fields.
Use when the workload needs 2k context and 20B parameters.
Use when the workload needs 2k context and 5B parameters.
Use when the workload needs 2k context and 128M parameters.
Use when the workload needs 2k context and 20B parameters.
Use when the workload needs 2k context and 3B parameters.
| Model | Use when | Released | Signals | Status |
|---|---|---|---|---|
| Palmyra Large | Use when the workload needs 2k context and 20B parameters. | 2023-03 | 2k context20B parameters | Current |
| Palmyra Base | Use when the workload needs 2k context and 5B parameters. | 2023-03 | 2k context5B parameters | Current |
| Palmyra Small | Use when the workload needs 2k context and 128M parameters. | 2023-03 | 2k context128M parameters | Current |
| Palmyra 20B | Use when the workload needs 2k context and 20B parameters. | 2023-03 | 2k context20B parameters | Current |
| Palmyra 3B | Use when the workload needs 2k context and 3B parameters. | 2023-03 | 2k context3B parameters | Current |
Release Timeline
1 release groupSpecifications(5 models)
| Model | Released | Context | Parameters |
|---|---|---|---|
| Palmyra Large | 2023-03 | 2k | 20B |
| Palmyra Base | 2023-03 | 2k | 5B |
| Palmyra Small | 2023-03 | 2k | 128M |
| Palmyra 20B | 2023-03 | 2k | 20B |
| Palmyra 3B | 2023-03 | 2k | 3B |
Frequently Asked Questions
- What is Palmyra used for?
- Palmyra is used for coding. The family description and listed model capabilities point to those workloads as the best fit.
- How does Palmyra compare to Camel?
- Palmyra by Writer is strongest where you need coding, while Camel by Writer is the closest related family to check for adjacent model selection. Palmyra has 5 listed variants and reaches up to 2k context, while Camel reaches up to 4k context, so compare the specs and pricing tables before choosing a production model.
- Which Palmyra model should I use?
- If price is the main constraint, use the pricing table first because Palmyra does not have complete provider pricing in the local data. For the most capable/latest local choice, evaluate Palmyra Large with 2k context.
