AI-ML
Ollama v0.21.3
RESUMEN
What's Changed api: accept "max" as a think value by @ParthSareen in openai: map responses reasoning effort to think by @ParthSareen in Full Changelog:
Descripción Detallada
What's Changed api: accept "max" as a think value by @ParthSareen in openai: map responses reasoning effort to think by @ParthSareen in Full Changelog:
Actualización de Ollama v0.21.3 con mejoras en la API y OpenAI.
- La API ahora acepta 'max' como un valor de pensamiento.
- Se mapean los esfuerzos de razonamiento de las respuestas de OpenAI al pensamiento.
A quién le importa
Todos los que usan Ollama y OpenAI.
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