Update README.md

README.md

@@ -73,41 +73,6 @@ MTP ON low-context peak	~70 tok/s
 
 These are local observations, not standardized benchmarks. Actual speed depends on prompt length, context length, oMLX version, generation settings, temperature, cache behavior, and Apple Silicon memory pressure.
 
-Intended use
-
-This model is intended for local oMLX / MLX inference, especially:
-
-* local agent workflows
-* coding assistant experiments
-* Runtime v2 / Claude Code style local agent tasks
-* long-context local reasoning and review
-* Apple Silicon local inference testing
-* oMLX native MTP experiments
-
-Not intended for
-
-* production use without your own testing
-* safety-critical decisions
-* use as an official Qwen/QwOpus release
-* non-MLX runtimes unless you know this format is supported
-
-This is not GGUF, AWQ, GPTQ, or bitsandbytes format. It is an MLX safetensors conversion.
-
-Suggested settings by task
-
-advisor / planning:
-  temperature: 0.3-0.4
-  top_p: 0.9-0.95
-execute / coding / shell edits:
-  temperature: 0.1-0.2
-  top_p: 0.9-0.95
-review:
-  temperature: 0.2-0.3
-  top_p: 0.9-0.95
-archive / summary:
-  temperature: 0.3-0.5
-  top_p: 0.9-0.95
-
 Known notes
 
 * Native MTP showed a measurable decode-speed improvement in local testing.