XVERSE-13B
13B multilingual base model with 8K context, free for commercial use.
About
XVERSE-13B is a multilingual large language model developed by Shenzhen Yuanxiang Technology (xverse-ai), with the 13B base model released August 7, 2023. It uses a standard decoder-only Transformer with an 8K context window, trained on over 3 trillion tokens spanning 40+ languages with strong Chinese and English performance.
The team positioned it as fully open source and free for commercial use, and followed up with an instruct-tuned chat variant, a 7B size, and a 256K-context version. Later XVERSE-13B-2 revisions expanded training to 3.2T tokens with function-calling support.
Specs
Context window
Pricing
Free quota
API access
Weights
Released
Benchmarks
Provider-reported unless a source says otherwise. Benchmarks are marketing-adjacent by nature — treat them as directional.
Use cases
Trust score — 7/10
Scored 7/10 (metrics tier, metrics-v1). Evidence: Verified via vendor site/docs/GitHub and reputable coverage, wave 132; Trending score 70/100 from researcher assessment.