SeaLLM Models by Alibaba
About
SeaLLMs is a family of large language models that focus on Southeast Asian (SEA) languages, addressing the linguistic bias present in many existing models that lean towards high-resource languages like English. By building on English-centric models and enhancing them with extended vocabulary and alignment tuning, SeaLLMs capture the nuances of SEA languages and respect local cultural norms, customs, and legalities. Notably, SeaLLMs perform exceptionally well across various linguistic tasks and even exceed ChatGPT-3.5's performance in several non-Latin SEA languages. The latest iteration, SeaLLMs-v3, sets itself apart with state-of-the-art capabilities in diverse tasks and improvements in trustworthiness and reduced hallucinations 26911.
Current Variants
Use-when guidance is derived from seed capabilities, context, release, and replacement fields.
Use when the workload needs 4k context and 13B parameters.
Use when the workload needs 32k context and 7B parameters.
| Model | Use when | Released | Signals | Status |
|---|---|---|---|---|
| SeaLLM 13B | Use when the workload needs 4k context and 13B parameters. | 2024-03 | 4k context13B parameters | Current |
| SeaLLM 7B | Use when the workload needs 32k context and 7B parameters. | 2024-03 | 32k context7B parameters | Current |
Release Timeline
1 release groupSpecifications(2 models)
| Model | Released | Context | Parameters |
|---|---|---|---|
| SeaLLM 13B | 2024-03 | 4k | 13B |
| SeaLLM 7B | 2024-03 | 32k | 7B |
Frequently Asked Questions
- What is SeaLLM used for?
- SeaLLM is used for math-heavy prompts and chatbot and role-playing use cases. The family description and listed model capabilities point to those workloads as the best fit.
- How does SeaLLM compare to Tongyi DeepResearch?
- SeaLLM by Alibaba is strongest where you need math-heavy prompts, while Tongyi DeepResearch by Alibaba is the closest related family to check for adjacent model selection. SeaLLM has 2 listed variants and reaches up to 32k context, while Tongyi DeepResearch reaches up to 131k context, so compare the specs and pricing tables before choosing a production model.
- Which SeaLLM model should I use?
- If price is the main constraint, use the pricing table first because SeaLLM does not have complete provider pricing in the local data. For the most capable/latest local choice, evaluate SeaLLM 7B with 32k context.
