GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
buddhist-nlp
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
buddhist-nlp
Model Tree
buddhist-nlp/mitra-qwen35-base-stage2
Input Modalities
Output Modalities
Supported Functionality
text
Insert the numbered tags into the target line, marking the word(s) that translate each numbered source word. Copy the target verbatim and omit numbers without a clear equivalent.SANSKRIT: <1>evam</1> <2>ca</2> <3>tat</3> <4>karaṇīyam</4>LEMMAS: <1>evam</1> <2>ca</2> <3>tad</3> <4>kṛ</4>TIBETAN: དེ་ནི་དེ་ལྟ་བུར་བྱ་སྟེ།OUTPUT:Output:
text
དེ་ནི་<1>དེ་ལྟ་བུར</1>་<4>བྱ</4>་སྟེ།The target line is TIBETAN:, ENGLISH: or CHINESE: (followed by a CHINESE_SEGMENTED: line with the word-segmented Chinese). Tibetan is given in Unicode. Use greedy decoding and stop at the first newline.
Full fine-tuning of MITRA Qwen3.5-9B (continued pretraining on Sanskrit, Tibetan, Buddhist Chinese and Pāli) for three epochs on 9,366 sentence pairs labelled by a commercial LLM. Directions are Sanskrit to Tibetan, Chinese and English, plus Chinese to Tibetan and English to Tibetan.
Link-level F1 against 1,489 held-out sentence pairs labelled by the same teacher: Sanskrit to Tibetan 88.9, Sanskrit to Chinese 89.2, Sanskrit to English 86.9. These figures measure agreement with the teacher, not correctness.
The weights use the Qwen3_5ForConditionalGeneration layout, which vLLM requires. The vision tower is unused.
Data and evaluation protocol: https://github.com/dharmamitra/dharmamitra-lexicon
Interactive lexicon: https://lexicon.dharmamitra.org