A 1.5B native tool-calling, Bash and Python fine-tune of Qwen2.5-Coder-1.5B-Instruct, shipped as merged bf16 weights and six GGUF quants. — Tool calling is the big gain. The base model’s tool-tag embeddings are untrained, so it barely emits its own format, and when asked for tools it mostly answers in Markdown JSON. Fixing that, plus the tool data, takes it from 52.7% to 86.2%. — Bash and Python improve without being traded away. NL2SH-ALFA gains 26 commands and Python gains 38 problems, both significant. MultiPL-E sh (writing whole shell programs) is unchanged. — InterCode needs a caveat.** 127 of its 200 tasks overlap NL2SH-ALFA’s test split, and the whole gain sits there. On the 73 independent tasks it is flat. Every model was run through the same harness in bf16, using its own chat template and native tool-call format. Notes: = the gap to mini-agent-v2 is statistically significant (both models answered the same questions; paired exact McNemar test, p < 0.05), so it is unlikely to be luck. No means the two are effectively tied. ** = that model was not tested on this benchmark (not a score of zero). GGUF metadata: architecture qwen2, 1,543,714,304 params, context 32768, EOS ,…
Модальности:
Генерация текста
Области применения:
Генерация кода Диалог / чат Вызов функций (Tool use)
Языки программирования:
Python Bash/Shell
Задача: Генерация текста
Автор: F-A-I-L
Теги: gguf, quantized, qwen2, qwen2.5-coder, lora, qlora, unsloth, trl
Лайков: 3 | Загрузок: 543
Описание основано на материалах HuggingFace. Перевод выполнен автоматически.
