AI-ML
Ollama v0.17.3
RESUMEN
What's Changed * Fixed issue where tool calls in the Qwen 3 and Qwen 3.5 model families would not be parsed correctly if emitted during thinking Full Changelog:
Descripción Detallada
What's Changed * Fixed issue where tool calls in the Qwen 3 and Qwen 3.5 model families would not be parsed correctly if emitted during thinking Full Changelog:
Se corrige un problema de análisis en los modelos Qwen 3 y 3.5.
- Se solucionó un problema donde las llamadas a herramientas en los modelos Qwen 3 y 3.5 no se analizaban correctamente durante el pensamiento.
A quién le importa
Todos los que usen los modelos Qwen 3 y 3.5.
Generado por IA · puede contener errores
Releases Relacionados
AI-ML
Ollama v0.32.5
## What's Changed * Fixed an MLX Metal bug that could reduce output quality for NVFP4 models, particularly Laguna. **Full Changelog**: https://github.com/ollama/ollama/compare/v0.32.4...v0.32.5
AI-ML
Ollama v0.32.4
## What's Changed - Support Laguna on Apple GPUs via the MLX engine - Quantize draft-model output heads at the requested type when creating speculative-decoding drafts. - Fixed Qwen3 MoE decoding for differently-quantized experts, plus faster packed gate/up projection (~4–9% on M5 Max). **Full
AI-ML