17 Downloads
Capabilities
Minimum system memory
Tags
17 Downloads
Capabilities
17 Downloads
Capabilities
Minimum system memory
Tags
17 Downloads
Capabilities
Minimum system memory
Tags
Last updated
Updated 25 days agobyREADME
MLX 4-bit and 6-bit conversions of medismera/Qwen3.8-27B-OBLITERATED-Mythos-Class-Agentic, with LM Studio reasoning controls.
This model definition links to the existing Hugging Face weights and exposes Reasoning Effort (Low, Medium, Extra High), Enable Thinking, and Preserve Thinking. Medium preserves the upstream Mythos template default. These are prompt-template controls, not a guaranteed reasoning-token budget.
Custom Fields
Special features defined by the model author
Reasoning Effort
: select
(default=medium)
Controls how much reasoning the model should perform.
Enable Thinking
: boolean
(default=true)
Controls whether the model will think before replying
Preserve Thinking
: boolean
(default=true)
Preserve reasoning content in all prior assistant turns instead of only the most recent one
Sources
The underlying model files this model uses
Minimum system memory
Tags
Last updated
Updated 25 days agobyREADME
MLX 4-bit and 6-bit conversions of medismera/Qwen3.8-27B-OBLITERATED-Mythos-Class-Agentic, with LM Studio reasoning controls.
This model definition links to the existing Hugging Face weights and exposes Reasoning Effort (Low, Medium, Extra High), Enable Thinking, and Preserve Thinking. Medium preserves the upstream Mythos template default. These are prompt-template controls, not a guaranteed reasoning-token budget.
Custom Fields
Special features defined by the model author
Reasoning Effort
: select
(default=medium)
Controls how much reasoning the model should perform.
Enable Thinking
: boolean
(default=true)
Controls whether the model will think before replying
Preserve Thinking
: boolean
(default=true)
Preserve reasoning content in all prior assistant turns instead of only the most recent one
Sources
The underlying model files this model uses
Both downloads retain vision weights. Image processing has not been tested. These are basic functionality checks, not a quality benchmark or evidence of improved agent performance. Memory requirements exceed weight-file sizes and depend on context and runtime.
lms get uzzero/qwen3.8-27b-mythos-mlx
Choose Qwen3.8 27B Mythos, publisher uzzero, in LM Studio. A raw Hugging Face checkpoint entry alone does not carry these custom controls.
Credit for the source model belongs to its creators. The conversions add quantization, packaging and initial checks, without additional training. See the linked Hugging Face model cards for provenance, test details and Apache-2.0 license notices.
Both downloads retain vision weights. Image processing has not been tested. These are basic functionality checks, not a quality benchmark or evidence of improved agent performance. Memory requirements exceed weight-file sizes and depend on context and runtime.
lms get uzzero/qwen3.8-27b-mythos-mlx
Choose Qwen3.8 27B Mythos, publisher uzzero, in LM Studio. A raw Hugging Face checkpoint entry alone does not carry these custom controls.
Credit for the source model belongs to its creators. The conversions add quantization, packaging and initial checks, without additional training. See the linked Hugging Face model cards for provenance, test details and Apache-2.0 license notices.