"Ripping your programming worries to shreds... fast... times TWO."
Tipping the scales at 42 layers and 717 tensors... the monster lives.
Two monsters in fact - in one.
This is MOE model, using V1 and V2 of Wolverine-Coder 11B which is a merge of two models noted below.
This repo has the source code for Version 2.
The MOE config gives you full access to both 11B models at full power - full 22B parameters (an additional shared expert brings this to 28B).
This MOE model generates stronger, more compact code with an enhanced understanding of your instructions and follows what you tell them to the letter.
Each 11B version is an overpowered - yet wickedly fast - CODING ENGINE are based on two of the best coder AIs:
"Qwen2.5-Coder-7B-Instruct"
[ https://huggingface.co/Qwen/Qwen2.5-Coder-7B-Instruct ]
and
"OlympicCoder-7B"
[ https://huggingface.co/open-r1/OlympicCoder-7B ]
(11Bs V1 and V2 are here: https://huggingface.co/DavidAU/Qwen2.5-Wolverine-CODER-11B-gguf/ )
These two models are stuffed into one compact powerhouse 11BX2 merge that is stronger in performance and understanding than both donor models and each of the 11B models created from these too.
There are TWO versions of this MOE model too (internal structure is different for each).
Quants Q3_K_M and Q6_K are available at the moment, of each version.
Limited GGUFS here:
https://huggingface.co/DavidAU/Qwen2.5-2X11B-CODER-Dueling-Wolverines-28B-gguf
More quants will show under "Quantizations" on the right as they become available.
NOTES:
Config / Settings
Model is set at 32k/32768 context for these GGUFS, full quants/full repos will be 128k/131072.
Requirements [Qwen 2.5 7B Coder default settings]:
Refer to either "Qwen2.5-Coder-7B-Instruct" and/or "OlympicCoder-7B" repos (above) for additional settings, benchmarks and usage.
Settings: CHAT / ROLEPLAY and/or SMOOTHER operation of this model:
In "KoboldCpp" or "oobabooga/text-generation-webui" or "Silly Tavern" ;
Set the "Smoothing_factor" to 1.5
: in KoboldCpp -> Settings->Samplers->Advanced-> "Smooth_F"
: in text-generation-webui -> parameters -> lower right.
: In Silly Tavern this is called: "Smoothing"
NOTE: For "text-generation-webui"
-> if using GGUFs you need to use "llama_HF" (which involves downloading some config files from the SOURCE version of this model)
Source versions (and config files) of my models are here:
OTHER OPTIONS:
Increase rep pen to 1.1 to 1.15 (you don't need to do this if you use "smoothing_factor")
If the interface/program you are using to run AI MODELS supports "Quadratic Sampling" ("smoothing") just make the adjustment as noted.
Highest Quality Settings / Optimal Operation Guide / Parameters and Samplers
This a "Class 1" model:
For all settings used for this model (including specifics for its "class"), including example generation(s) and for advanced settings guide (which many times addresses any model issue(s)), including methods to improve model performance for all use case(s) as well as chat, roleplay and other use case(s) please see:
You can see all parameters used for generation, in addition to advanced parameters and samplers to get the most out of this model here:
