zelk12

GGUF Static: https://huggingface.co/mradermacher/MT4-Gen2-gemma-2-9B-GGUF This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT4-Gen2-IMM-gemma-2-9B zelk12/MT4-Gen2-GBMAMU-gemma-2-9B The following YAML configuration was used to produce this model: Open LLM Leaderboard Evaluation Results Detailed results can be found here | Metric |Value| |-------------------|----:| |Avg. |33.79| |IFEval (0-Shot) |80.51| |BBH (3-Shot) |44.18| |MATH Lvl 5 (4-Shot)|15.71| |GPQA (0-shot) |12.75| |MuSR (0-shot) |12.21| |MMLU-PRO (5-shot) |37.42|

zelk12/MT2-Gen2gemma-3-12B-Q6K-GGUF This model was converted to GGUF format from `zelk12/MT2-Gen2gemma-3-12B` using llama.cpp via the ggml.ai's GGUF-my-repo space. Refer to the original model card for more details on the model. Use with llama.cpp Install llama.cpp through brew (works on Mac and Linux) Note: You can also use this checkpoint directly through the usage steps listed in the Llama.cpp repo as well. Step 2: Move into the llama.cpp folder and build it with `LLAMACURL=1` flag along with other hardware-specific flags (for ex: LLAMACUDA=1 for Nvidia GPUs on Linux).

zelk12/MT1-gemma-3-12B-Q6K-GGUF This model was converted to GGUF format from `zelk12/MT1-gemma-3-12B` using llama.cpp via the ggml.ai's GGUF-my-repo space. Refer to the original model card for more details on the model. Use with llama.cpp Install llama.cpp through brew (works on Mac and Linux) Note: You can also use this checkpoint directly through the usage steps listed in the Llama.cpp repo as well. Step 2: Move into the llama.cpp folder and build it with `LLAMACURL=1` flag along with other hardware-specific flags (for ex: LLAMACUDA=1 for Nvidia GPUs on Linux).

zelk12/MT5-Gen3gemma-3-12B-Q6K-GGUF This model was converted to GGUF format from `zelk12/MT5-Gen3gemma-3-12B` using llama.cpp via the ggml.ai's GGUF-my-repo space. Refer to the original model card for more details on the model. Use with llama.cpp Install llama.cpp through brew (works on Mac and Linux) Note: You can also use this checkpoint directly through the usage steps listed in the Llama.cpp repo as well. Step 2: Move into the llama.cpp folder and build it with `LLAMACURL=1` flag along with other hardware-specific flags (for ex: LLAMACUDA=1 for Nvidia GPUs on Linux).

zelk12/26052025TestLazyMergekitgemma-3-12B-Q6K-GGUF This model was converted to GGUF format from `zelk12/26052025TestLazyMergekitgemma-3-12B` using llama.cpp via the ggml.ai's GGUF-my-repo space. Refer to the original model card for more details on the model. Use with llama.cpp Install llama.cpp through brew (works on Mac and Linux) Note: You can also use this checkpoint directly through the usage steps listed in the Llama.cpp repo as well. Step 2: Move into the llama.cpp folder and build it with `LLAMACURL=1` flag along with other hardware-specific flags (for ex: LLAMACUDA=1 for Nvidia GPUs on Linux).

zelk12/MT4-Gen3gemma-3-12B-Q6K-GGUF This model was converted to GGUF format from `zelk12/MT4-Gen3gemma-3-12B` using llama.cpp via the ggml.ai's GGUF-my-repo space. Refer to the original model card for more details on the model. Use with llama.cpp Install llama.cpp through brew (works on Mac and Linux) Note: You can also use this checkpoint directly through the usage steps listed in the Llama.cpp repo as well. Step 2: Move into the llama.cpp folder and build it with `LLAMACURL=1` flag along with other hardware-specific flags (for ex: LLAMACUDA=1 for Nvidia GPUs on Linux).

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT2-Gen4-IMU-gemma-2-9B zelk12/MT2-Gen4-MAGBMM-gemma-2-9B The following YAML configuration was used to produce this model: Open LLM Leaderboard Evaluation Results Detailed results can be found here | Metric |Value| |-------------------|----:| |Avg. |31.86| |IFEval (0-Shot) |78.96| |BBH (3-Shot) |43.78| |MATH Lvl 5 (4-Shot)| 8.31| |GPQA (0-shot) |12.75| |MuSR (0-shot) |10.47| |MMLU-PRO (5-shot) |36.90|

zelk12/MTMMe-Merge-gemma-2-9BNuSLERP-Q6K-GGUF This model was converted to GGUF format from `zelk12/MTMMe-Merge-gemma-2-9BNuSLERP` using llama.cpp via the ggml.ai's GGUF-my-repo space. Refer to the original model card for more details on the model. Use with llama.cpp Install llama.cpp through brew (works on Mac and Linux) Note: You can also use this checkpoint directly through the usage steps listed in the Llama.cpp repo as well. Step 2: Move into the llama.cpp folder and build it with `LLAMACURL=1` flag along with other hardware-specific flags (for ex: LLAMACUDA=1 for Nvidia GPUs on Linux).

zelk12/MT2-Gen1-gemma-2-9B-Q6K-GGUF This model was converted to GGUF format from `zelk12/MT2-Gen1-gemma-2-9B` using llama.cpp via the ggml.ai's GGUF-my-repo space. Refer to the original model card for more details on the model. Use with llama.cpp Install llama.cpp through brew (works on Mac and Linux) Note: You can also use this checkpoint directly through the usage steps listed in the Llama.cpp repo as well. Step 2: Move into the llama.cpp folder and build it with `LLAMACURL=1` flag along with other hardware-specific flags (for ex: LLAMACUDA=1 for Nvidia GPUs on Linux).

zelk12/MT-Merge2-gemma-2-9B-Q6K-GGUF This model was converted to GGUF format from `zelk12/MT-Merge2-gemma-2-9B` using llama.cpp via the ggml.ai's GGUF-my-repo space. Refer to the original model card for more details on the model. Use with llama.cpp Install llama.cpp through brew (works on Mac and Linux) Note: You can also use this checkpoint directly through the usage steps listed in the Llama.cpp repo as well. Step 2: Move into the llama.cpp folder and build it with `LLAMACURL=1` flag along with other hardware-specific flags (for ex: LLAMACUDA=1 for Nvidia GPUs on Linux).

GGUF Static: https://huggingface.co/mradermacher/MT2-Gen3-gemma-2-9B-GGUF This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT2-Gen3-BMAMUG-gemma-2-9B zelk12/MT2-Gen3-IMM-gemma-2-9B The following YAML configuration was used to produce this model: Open LLM Leaderboard Evaluation Results Detailed results can be found here | Metric |Value| |-------------------|----:| |Avg. |32.97| |IFEval (0-Shot) |78.10| |BBH (3-Shot) |44.01| |MATH Lvl 5 (4-Shot)|13.29| |GPQA (0-shot) |12.86| |MuSR (0-shot) |12.05| |MMLU-PRO (5-shot) |37.49|

zelk12/MT-gemma-3-12B-Q6K-GGUF This model was converted to GGUF format from `zelk12/MT-gemma-3-12B` using llama.cpp via the ggml.ai's GGUF-my-repo space. Refer to the original model card for more details on the model. Use with llama.cpp Install llama.cpp through brew (works on Mac and Linux) Note: You can also use this checkpoint directly through the usage steps listed in the Llama.cpp repo as well. Step 2: Move into the llama.cpp folder and build it with `LLAMACURL=1` flag along with other hardware-specific flags (for ex: LLAMACUDA=1 for Nvidia GPUs on Linux).

GGUF Static: https://huggingface.co/mradermacher/MT3-Gen2-gemma-2-9B-GGUF This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT3-Gen2-GMM-gemma-2-9B zelk12/MT3-Gen2-IMUBMA-gemma-2-9B The following YAML configuration was used to produce this model: Open LLM Leaderboard Evaluation Results Detailed results can be found here | Metric |Value| |-------------------|----:| |Avg. |30.96| |IFEval (0-Shot) |78.43| |BBH (3-Shot) |43.94| |MATH Lvl 5 (4-Shot)| 2.04| |GPQA (0-shot) |14.32| |MuSR (0-shot) |10.02| |MMLU-PRO (5-shot) |37.03|

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MTM-Merge-MA-gemma-2-MTM4MTM5-9B zelk12/MTM-Merge-GP-gemma-2-MTM5MTM4-9B The following YAML configuration was used to produce this model:

zelk12/MT-Gen8-gemma-2-9B-Q6K-GGUF This model was converted to GGUF format from `zelk12/MT-Gen8-gemma-2-9B` using llama.cpp via the ggml.ai's GGUF-my-repo space. Refer to the original model card for more details on the model. Use with llama.cpp Install llama.cpp through brew (works on Mac and Linux) Note: You can also use this checkpoint directly through the usage steps listed in the Llama.cpp repo as well. Step 2: Move into the llama.cpp folder and build it with `LLAMACURL=1` flag along with other hardware-specific flags (for ex: LLAMACUDA=1 for Nvidia GPUs on Linux).

zelk12/MT-Gen11-gemma-2-9B-Q6K-GGUF This model was converted to GGUF format from `zelk12/MT-Gen11-gemma-2-9B` using llama.cpp via the ggml.ai's GGUF-my-repo space. Refer to the original model card for more details on the model. Use with llama.cpp Install llama.cpp through brew (works on Mac and Linux) Note: You can also use this checkpoint directly through the usage steps listed in the Llama.cpp repo as well. Step 2: Move into the llama.cpp folder and build it with `LLAMACURL=1` flag along with other hardware-specific flags (for ex: LLAMACUDA=1 for Nvidia GPUs on Linux).

zelk12/MT7-Gen3gemma-3-12B-Q6K-GGUF This model was converted to GGUF format from `zelk12/MT7-Gen3gemma-3-12B` using llama.cpp via the ggml.ai's GGUF-my-repo space. Refer to the original model card for more details on the model. Use with llama.cpp Install llama.cpp through brew (works on Mac and Linux) Note: You can also use this checkpoint directly through the usage steps listed in the Llama.cpp repo as well. Step 2: Move into the llama.cpp folder and build it with `LLAMACURL=1` flag along with other hardware-specific flags (for ex: LLAMACUDA=1 for Nvidia GPUs on Linux).

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT2-Gen1-MMMUMAG-gemma-2-9B zelk12/MT2-Gen1-IB-gemma-2-9B The following YAML configuration was used to produce this model: Open LLM Leaderboard Evaluation Results Detailed results can be found here | Metric |Value| |-------------------|----:| |Avg. |32.46| |IFEval (0-Shot) |78.56| |BBH (3-Shot) |44.14| |MATH Lvl 5 (4-Shot)|10.12| |GPQA (0-shot) |12.42| |MuSR (0-shot) |12.01| |MMLU-PRO (5-shot) |37.52|

MT4-Gen2gemma-3-12B is a merge of the following models using LazyMergekit: zelk12/MT-Gen1-gemma-3-12B soob3123/amoral-gemma3-12B-v2 zelk12/MT1-gemma-3-12B IlyaGusev/saigagemma312b TheDrummer/Fallen-Gemma3-12B-v1

GGUF Static: https://huggingface.co/mradermacher/MT2-Gen2-gemma-2-9B-GGUF This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT2-Gen2-BGMAMU-gemma-2-9B zelk12/MT2-Gen2-IMM-gemma-2-9B The following YAML configuration was used to produce this model: Open LLM Leaderboard Evaluation Results Detailed results can be found here | Metric |Value| |-------------------|----:| |Avg. |33.47| |IFEval (0-Shot) |78.89| |BBH (3-Shot) |44.04| |MATH Lvl 5 (4-Shot)|14.80| |GPQA (0-shot) |12.86| |MuSR (0-shot) |12.58| |MMLU-PRO (5-shot) |37.65|

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT-Gen6fix-UW-gemma-2-9B zelk12/MT-Gen6fix-CN-gemma-2-9B The following YAML configuration was used to produce this model:

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT-Gen8-CU-gemma-2-9B zelk12/MT-Gen8-WN-gemma-2-9B The following YAML configuration was used to produce this model:

This is a merge of pre-trained language models created using mergekit. This model was merged using the DARE TIES merge method using soob3123/amoral-gemma3-12B-v2 as a base. The following models were included in the merge: IlyaGusev/saigagemma312b The following YAML configuration was used to produce this model:

GGUF Static: https://huggingface.co/models?other=basemodel:quantized:zelk12/MT2-Gen2-BG-gemma-2-9B This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT2-Gen2-BB-gemma-2-MTMMT5-9B zelk12/MT2-Gen2-GP-gemma-2-RIv0.1MTM-9B The following YAML configuration was used to produce this model:

GGUF Static: https://huggingface.co/mradermacher/MT2-Gen3-IF-gemma-2-MT4g2S2-9B-GGUF This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: allknowingroger/Gemmaslerp2-9B zelk12/MT4-Gen2-gemma-2-9B The following YAML configuration was used to produce this model:

GGUF Static: https://huggingface.co/mradermacher/MT3-Gen3-BB-gemma-2-MT2RIv0.1-9B-GGUF This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/recoilme-gemma-2-Ifable-9B-v0.1 zelk12/MT2-gemma-2-9B The following YAML configuration was used to produce this model:

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT-Merge-gemma-2-9B recoilme/recoilme-gemma-2-9B-v0.4 The following YAML configuration was used to produce this model:

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT3-Gen4-GP-gemma-2-MT5g3MT2g2-9B zelk12/MT3-Gen4-BB-gemma-2-MT2g2MT5g3-9B The following YAML configuration was used to produce this model:

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT3-Gen4-IF-gemma-2-Riv0.1RAv0.1-9B zelk12/MT3-Gen4-MU-gemma-2-Sv1IBT-9B The following YAML configuration was used to produce this model:

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT1-Gen5-BMMIMU-gemma-2-9B zelk12/MT1-Gen5-MAG-gemma-2-9B The following YAML configuration was used to produce this model:

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: djuna/Gemma-2-gemmama-9b IlyaGusev/gemma-2-9b-it-abliterated The following YAML configuration was used to produce this model:

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT2-Gen7-CN-gemma-2-9B zelk12/MT2-Gen7-UW-gemma-2-9B The following YAML configuration was used to produce this model:

MT3-Gen2gemma-3-12B is a merge of the following models using LazyMergekit: zelk12/MT-Gen1-gemma-3-12B soob3123/amoral-gemma3-12B-v2 zelk12/MT1-gemma-3-12B IlyaGusev/saigagemma312b TheDrummer/Fallen-Gemma3-12B-v1

zelk12/MT5-Gen1-gemma-2-9B-Q6K-GGUF This model was converted to GGUF format from `zelk12/MT5-Gen1-gemma-2-9B` using llama.cpp via the ggml.ai's GGUF-my-repo space. Refer to the original model card for more details on the model. Use with llama.cpp Install llama.cpp through brew (works on Mac and Linux) Note: You can also use this checkpoint directly through the usage steps listed in the Llama.cpp repo as well. Step 2: Move into the llama.cpp folder and build it with `LLAMACURL=1` flag along with other hardware-specific flags (for ex: LLAMACUDA=1 for Nvidia GPUs on Linux).

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT-Merge-gemma-2-9B zelk12/recoilme-gemma-2-Ifable-9B-v0.1 The following YAML configuration was used to produce this model:

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT5-Gen4-IMUBMA-gemma-2-9B zelk12/MT5-Gen4-MMG-gemma-2-9B The following YAML configuration was used to produce this model: Open LLM Leaderboard Evaluation Results Detailed results can be found here | Metric |Value| |-------------------|----:| |Avg. |33.77| |IFEval (0-Shot) |78.35| |BBH (3-Shot) |44.32| |MATH Lvl 5 (4-Shot)|17.07| |GPQA (0-shot) |13.76| |MuSR (0-shot) |11.35| |MMLU-PRO (5-shot) |37.74|

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT2-Gen5-IF-gemma-2-S2MT4g2-9B zelk12/MT2-Gen5-MA-gemma-2-S2MT3g4-9B The following YAML configuration was used to produce this model:

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT2-Gen5-GBMMMU-gemma-2-9B zelk12/MT2-Gen5-MAI-gemma-2-9B The following YAML configuration was used to produce this model:

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: allknowingroger/Gemmaslerp2-9B zelk12/T31122024203920-gemma-2-9B The following YAML configuration was used to produce this model:

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT1-Gen7-W-gemma-2-MTg6MTg6f-9B zelk12/MT1-Gen7-U-gemma-2-MTg6fMTg6-9B The following YAML configuration was used to produce this model:

zelk12/28052025Test2LazyMergekitgemma-3-12B-Q6K-GGUF This model was converted to GGUF format from `zelk12/28052025Test2LazyMergekitgemma-3-12B` using llama.cpp via the ggml.ai's GGUF-my-repo space. Refer to the original model card for more details on the model. Use with llama.cpp Install llama.cpp through brew (works on Mac and Linux) Note: You can also use this checkpoint directly through the usage steps listed in the Llama.cpp repo as well. Step 2: Move into the llama.cpp folder and build it with `LLAMACURL=1` flag along with other hardware-specific flags (for ex: LLAMACUDA=1 for Nvidia GPUs on Linux).

MT8-Gen2gemma-3-12B is a merge of the following models using LazyMergekit: zelk12/MT-Gen1-gemma-3-12B soob3123/amoral-gemma3-12B-v2 zelk12/MT1-gemma-3-12B IlyaGusev/saigagemma312b TheDrummer/Fallen-Gemma3-12B-v1

zelk12/MT2-Gen3gemma-3-12B-Q6K-GGUF This model was converted to GGUF format from `zelk12/MT2-Gen3gemma-3-12B` using llama.cpp via the ggml.ai's GGUF-my-repo space. Refer to the original model card for more details on the model. Use with llama.cpp Install llama.cpp through brew (works on Mac and Linux) Note: You can also use this checkpoint directly through the usage steps listed in the Llama.cpp repo as well. Step 2: Move into the llama.cpp folder and build it with `LLAMACURL=1` flag along with other hardware-specific flags (for ex: LLAMACUDA=1 for Nvidia GPUs on Linux).

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT3-Gen4-MAMM-gemma-2-9B zelk12/MT3-Gen4-GBMUI-gemma-2-9B The following YAML configuration was used to produce this model: Open LLM Leaderboard Evaluation Results Detailed results can be found here | Metric |Value| |-------------------|----:| |Avg. |34.49| |IFEval (0-Shot) |77.37| |BBH (3-Shot) |43.78| |MATH Lvl 5 (4-Shot)|20.47| |GPQA (0-shot) |12.98| |MuSR (0-shot) |14.72| |MMLU-PRO (5-shot) |37.64|

This is a merge of pre-trained language models created using mergekit. This model was merged using the DARE TIES merge method using zelk12/MT-Gen6fix-gemma-2-9B as a base. The following models were included in the merge: TheDrummer/Tiger-Gemma-9B-v3 IlyaGusev/gemma-2-9b-it-abliterated zelk12/MT1-Gen7-gemma-2-9B Sorawiz/Gemma-9B-Chat zelk12/MT-Merge6-gemma-2-9B The following YAML configuration was used to produce this model:

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: recoilme/recoilme-gemma-2-9B-v0.4 lemon07r/Gemma-2-Ataraxy-v2-9B The following YAML configuration was used to produce this model: Open LLM Leaderboard Evaluation Results Detailed results can be found here | Metric |Value| |-------------------|----:| |Avg. |28.42| |IFEval (0-Shot) |72.08| |BBH (3-Shot) |42.49| |MATH Lvl 5 (4-Shot)| 0.00| |GPQA (0-shot) |13.31| |MuSR (0-shot) | 7.76| |MMLU-PRO (5-shot) |34.90|

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT3-Gen5-MAI-gemma-2-9Bv1 zelk12/MT3-Gen5-MMGBMU-gemma-2-9Bv1 The following YAML configuration was used to produce this model:

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MTM-Merge-MMMUBI-gemma-2-9B zelk12/MTM-Merge-GMA-gemma-2-9B The following YAML configuration was used to produce this model:

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: recoilme/recoilme-gemma-2-9B-v0.4 ehristoforu/Gemma2-9B-it-psy10k-mentalhealth The following YAML configuration was used to produce this model: Open LLM Leaderboard Evaluation Results Detailed results can be found here | Metric |Value| |-------------------|----:| |Avg. |32.18| |IFEval (0-Shot) |74.45| |BBH (3-Shot) |42.13| |MATH Lvl 5 (4-Shot)|16.47| |GPQA (0-shot) |12.53| |MuSR (0-shot) |12.18| |MMLU-PRO (5-shot) |35.34|

GGUF Static: https://huggingface.co/mradermacher/MT4-Gen2-MM-gemma-2-Rv0.4MT1-9B-GGUF This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT1-gemma-2-9B recoilme/recoilme-gemma-2-9B-v0.4 The following YAML configuration was used to produce this model:

GGUF Static: https://huggingface.co/mradermacher/MT5-Gen2-gemma-2-9B-GGUF This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT5-Gen2-BI-gemma-2-9B zelk12/MT5-Gen2-MMGMAMU-gemma-2-9B The following YAML configuration was used to produce this model: Open LLM Leaderboard Evaluation Results Detailed results can be found here | Metric |Value| |-------------------|----:| |Avg. |32.60| |IFEval (0-Shot) |79.62| |BBH (3-Shot) |44.11| |MATH Lvl 5 (4-Shot)|10.35| |GPQA (0-shot) |13.53| |MuSR (0-shot) |10.44| |MMLU-PRO (5-shot) |37.55|

GGUF Static: https://huggingface.co/mradermacher/MT-Gen3-MU-gemma-2-Sv1IBTMTg2-9B-GGUF This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: gmonsoon/SahabatAI-Lion-9B-TIES-v1 zelk12/MT-Gen2-gemma-2-9B The following YAML configuration was used to produce this model:

GGUF Static: https://huggingface.co/mradermacher/MT1-Gen3-MM-gemma-2-MTg2S2-9B-GGUF This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: allknowingroger/Gemmaslerp2-9B zelk12/MT-Gen2-gemma-2-9B The following YAML configuration was used to produce this model:

GGUF Static: https://huggingface.co/mradermacher/MT3-Gen3-GMAMUB-gemma-2-9B-GGUF This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT3-Gen3-MUB-gemma-2-9B zelk12/MT3-Gen3-GMA-gemma-2-9B The following YAML configuration was used to produce this model:

GGUF Static: https://huggingface.co/mradermacher/MT-Gen4-MU-gemma-2-Sv1IBTS5-9B-GGUF This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: allknowingroger/GemmaSlerp5-10B gmonsoon/SahabatAI-Lion-9B-TIES-v1 The following YAML configuration was used to produce this model:

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: nhyha/N3Ngemma-2-9b-it202410291532 allknowingroger/Gemmaslerp2-9B The following YAML configuration was used to produce this model:

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT-Gen5-MUMMIG-gemma-2-9B zelk12/MT-Gen5-BMA-gemma-2-9B The following YAML configuration was used to produce this model:

zelk12/MT3-Gen5-gemma-2-9Bv1-Q6K-GGUF This model was converted to GGUF format from `zelk12/MT3-Gen5-gemma-2-9Bv1` using llama.cpp via the ggml.ai's GGUF-my-repo space. Refer to the original model card for more details on the model. Use with llama.cpp Install llama.cpp through brew (works on Mac and Linux) Note: You can also use this checkpoint directly through the usage steps listed in the Llama.cpp repo as well. Step 2: Move into the llama.cpp folder and build it with `LLAMACURL=1` flag along with other hardware-specific flags (for ex: LLAMACUDA=1 for Nvidia GPUs on Linux).

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT-Merge2-MU-gemma-2-MTg2MT1g2-9B zelk12/MT1-Gen2-gemma-2-9B The following YAML configuration was used to produce this model:

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT-Gen6-NC-gemma-2-9B zelk12/MT-Gen6-UW-gemma-2-9B The following YAML configuration was used to produce this model:

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: TheDrummer/Tiger-Gemma-9B-v3 IlyaGusev/gemma-2-9b-it-abliterated The following YAML configuration was used to produce this model:

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT3-Gen6-UC-gemma-2-9B zelk12/MT3-Gen6-WN-gemma-2-9B The following YAML configuration was used to produce this model:

zelk12/MT-Merge7-gemma-2-9B-Q6K-GGUF This model was converted to GGUF format from `zelk12/MT-Merge7-gemma-2-9B` using llama.cpp via the ggml.ai's GGUF-my-repo space. Refer to the original model card for more details on the model. Use with llama.cpp Install llama.cpp through brew (works on Mac and Linux) Note: You can also use this checkpoint directly through the usage steps listed in the Llama.cpp repo as well. Step 2: Move into the llama.cpp folder and build it with `LLAMACURL=1` flag along with other hardware-specific flags (for ex: LLAMACUDA=1 for Nvidia GPUs on Linux).

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: ifable/gemma-2-Ifable-9B recoilme/recoilme-gemma-2-9B-v0.4 The following YAML configuration was used to produce this model: Open LLM Leaderboard Evaluation Results Detailed results can be found here | Metric |Value| |-------------------|----:| |Avg. |32.05| |IFEval (0-Shot) |79.44| |BBH (3-Shot) |43.39| |MATH Lvl 5 (4-Shot)| 7.93| |GPQA (0-shot) |13.53| |MuSR (0-shot) |11.10| |MMLU-PRO (5-shot) |36.93|

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: recoilme/recoilme-gemma-2-9B-v0.4 zelk12/MT1-gemma-2-9B The following YAML configuration was used to produce this model:

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT5-Gen1-MA-gemma-2-Av4cMT1-9B zelk12/MT5-Gen1-MU-gemma-2-GMT1-9B The following YAML configuration was used to produce this model:

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT5-Gen1-MMGMUMA-gemma-2-9B zelk12/MT5-Gen1-BI-gemma-2-9B The following YAML configuration was used to produce this model: Open LLM Leaderboard Evaluation Results Detailed results can be found here | Metric |Value| |-------------------|----:| |Avg. |31.90| |IFEval (0-Shot) |78.31| |BBH (3-Shot) |44.18| |MATH Lvl 5 (4-Shot)| 6.87| |GPQA (0-shot) |12.98| |MuSR (0-shot) |11.61| |MMLU-PRO (5-shot) |37.43|

GGUF Static: https://huggingface.co/mradermacher/MT-Gen2-MUB-gemma-2-9B-GGUF This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT-Gen2-MU-gemma-2-MT1RAv0.1t0.25-9B zelk12/MT-Gen2-BB-gemma-2-MTMMT2-9B The following YAML configuration was used to produce this model:

GGUF Static: https://huggingface.co/mradermacher/MT3-Gen3-MA-gemma-2-S4MT2-9B-GGUF This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT2-gemma-2-9B allknowingroger/Gemmaslerp4-10B The following YAML configuration was used to produce this model:

GGUF Static: https://huggingface.co/mradermacher/MT3-Gen3-MU-gemma-2-Rv0.3MTM2MU-9B-GGUF This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT-Merge2-MU-gemma-2-MTg2MT1g2-9B recoilme/recoilme-gemma-2-9B-v0.3 The following YAML configuration was used to produce this model:

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT3-Gen5-BMU-gemma-2-9B zelk12/MT3-Gen5-MMG-gemma-2-9B The following YAML configuration was used to produce this model:

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT-Merge4-gemma-2-9B zelk12/MT1-Gen2-gemma-2-9B The following YAML configuration was used to produce this model:

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: recoilme/recoilme-gemma-2-9B-v0.2 djuna/Gemma-2-gemmama-9b The following YAML configuration was used to produce this model:

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT-Gen6fix-gemma-2-9B djuna/Gemma-2-gemmama-9b The following YAML configuration was used to produce this model:

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: Sorawiz/Gemma-9B-Chat IlyaGusev/gemma-2-9b-it-abliterated The following YAML configuration was used to produce this model:

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT1-Gen8-gemma-2-9B zelk12/MT-Gen8-gemma-2-9B The following YAML configuration was used to produce this model:

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT3-Gen9-U-gemma-2-MT2g6MTg7-9B zelk12/MT3-Gen9-W-gemma-2-MTg7MT2g6-9B The following YAML configuration was used to produce this model:

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT3-Gen9-N-gemma-2-MTMaMe02012025163610Gv1It-9B zelk12/MT3-Gen9-C-gemma-2-Gv1ItMT2g6-9B The following YAML configuration was used to produce this model:

zelk12/gemma3-vikhr-4b-Q6K-GGUF This model was converted to GGUF format from `Vikhrmodels/gemma3-vikhr-4b` using llama.cpp via the ggml.ai's GGUF-my-repo space. Refer to the original model card for more details on the model. Use with llama.cpp Install llama.cpp through brew (works on Mac and Linux) Note: You can also use this checkpoint directly through the usage steps listed in the Llama.cpp repo as well. Step 2: Move into the llama.cpp folder and build it with `LLAMACURL=1` flag along with other hardware-specific flags (for ex: LLAMACUDA=1 for Nvidia GPUs on Linux).

zelk12/MT2-gemma-3-12B-Q6K-GGUF This model was converted to GGUF format from `zelk12/MT2-gemma-3-12B` using llama.cpp via the ggml.ai's GGUF-my-repo space. Refer to the original model card for more details on the model. Use with llama.cpp Install llama.cpp through brew (works on Mac and Linux) Note: You can also use this checkpoint directly through the usage steps listed in the Llama.cpp repo as well. Step 2: Move into the llama.cpp folder and build it with `LLAMACURL=1` flag along with other hardware-specific flags (for ex: LLAMACUDA=1 for Nvidia GPUs on Linux).

zelk12/MT-Gen1-gemma-3-12B-Q6K-GGUF This model was converted to GGUF format from `zelk12/MT-Gen1-gemma-3-12B` using llama.cpp via the ggml.ai's GGUF-my-repo space. Refer to the original model card for more details on the model. Use with llama.cpp Install llama.cpp through brew (works on Mac and Linux) Note: You can also use this checkpoint directly through the usage steps listed in the Llama.cpp repo as well. Step 2: Move into the llama.cpp folder and build it with `LLAMACURL=1` flag along with other hardware-specific flags (for ex: LLAMACUDA=1 for Nvidia GPUs on Linux).

zelk12/26052025Test1LazyMergekitgemma-3-12B-Q6K-GGUF This model was converted to GGUF format from `zelk12/26052025Test1LazyMergekitgemma-3-12B` using llama.cpp via the ggml.ai's GGUF-my-repo space. Refer to the original model card for more details on the model. Use with llama.cpp Install llama.cpp through brew (works on Mac and Linux) Note: You can also use this checkpoint directly through the usage steps listed in the Llama.cpp repo as well. Step 2: Move into the llama.cpp folder and build it with `LLAMACURL=1` flag along with other hardware-specific flags (for ex: LLAMACUDA=1 for Nvidia GPUs on Linux).

zelk12/MT6-Gen3gemma-3-12B-Q6K-GGUF This model was converted to GGUF format from `zelk12/MT6-Gen3gemma-3-12B` using llama.cpp via the ggml.ai's GGUF-my-repo space. Refer to the original model card for more details on the model. Use with llama.cpp Install llama.cpp through brew (works on Mac and Linux) Note: You can also use this checkpoint directly through the usage steps listed in the Llama.cpp repo as well. Step 2: Move into the llama.cpp folder and build it with `LLAMACURL=1` flag along with other hardware-specific flags (for ex: LLAMACUDA=1 for Nvidia GPUs on Linux).

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: lemon07r/Gemma-2-Ataraxy-v2-9B recoilme/recoilme-gemma-2-9B-v0.4 The following YAML configuration was used to produce this model: Open LLM Leaderboard Evaluation Results Detailed results can be found here | Metric |Value| |-------------------|----:| |Avg. |30.33| |IFEval (0-Shot) |76.49| |BBH (3-Shot) |43.71| |MATH Lvl 5 (4-Shot)| 1.28| |GPQA (0-shot) |13.31| |MuSR (0-shot) |10.30| |MMLU-PRO (5-shot) |36.90|

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT1-GB-gemma-2-9B zelk12/MT1-IMMMU-gemma-2-9B The following YAML configuration was used to produce this model: Open LLM Leaderboard Evaluation Results Detailed results can be found here | Metric |Value| |-------------------|----:| |Avg. |33.37| |IFEval (0-Shot) |79.47| |BBH (3-Shot) |44.16| |MATH Lvl 5 (4-Shot)|13.37| |GPQA (0-shot) |12.75| |MuSR (0-shot) |13.16| |MMLU-PRO (5-shot) |37.31|

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT4-Gen5-GMA-gemma-2-9B zelk12/MT4-Gen5-IBMUMM-gemma-2-9B The following YAML configuration was used to produce this model:

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT5-Gen5-GI-gemma-2-9B zelk12/MT5-Gen5-MABMUMM-gemma-2-9B The following YAML configuration was used to produce this model:

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT-Gen7-CN-gemma-2-9B zelk12/MT-Gen7-UW-gemma-2-9B The following YAML configuration was used to produce this model:

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT1-Gen7-UW-gemma-2-9B zelk12/MT1-Gen7-CN-gemma-2-9B The following YAML configuration was used to produce this model:

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: recoilme/recoilme-gemma-2-9B-v0.4 lemon07r/Gemma-2-Ataraxy-v2-9B The following YAML configuration was used to produce this model: Open LLM Leaderboard Evaluation Results Detailed results can be found here | Metric |Value| |-------------------|----:| |Avg. |33.06| |IFEval (0-Shot) |77.07| |BBH (3-Shot) |43.85| |MATH Lvl 5 (4-Shot)|14.12| |GPQA (0-shot) |12.42| |MuSR (0-shot) |13.13| |MMLU-PRO (5-shot) |37.78|

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: recoilme/recoilme-gemma-2-9B-v0.4 nbeerbower/Gemma2-Gutenberg-Doppel-9B The following YAML configuration was used to produce this model: Open LLM Leaderboard Evaluation Results Detailed results can be found here | Metric |Value| |-------------------|----:| |Avg. |31.46| |IFEval (0-Shot) |76.15| |BBH (3-Shot) |43.94| |MATH Lvl 5 (4-Shot)| 6.34| |GPQA (0-shot) |12.19| |MuSR (0-shot) |13.31| |MMLU-PRO (5-shot) |36.84|

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/recoilme-gemma-2-Ataraxy-9B-v0.1-t0.75 zelk12/recoilme-gemma-2-Ataraxy-9B-v0.1-t0.25 The following YAML configuration was used to produce this model: Open LLM Leaderboard Evaluation Results Detailed results can be found here | Metric |Value| |-------------------|----:| |Avg. |30.10| |IFEval (0-Shot) |76.00| |BBH (3-Shot) |43.63| |MATH Lvl 5 (4-Shot)| 1.13| |GPQA (0-shot) |13.09| |MuSR (0-shot) | 9.84| |MMLU-PRO (5-shot) |36.92|

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT2-MMMAGMU-gemma-2-9B zelk12/MT2-IB-gemma-2-9B The following YAML configuration was used to produce this model: Open LLM Leaderboard Evaluation Results Detailed results can be found here | Metric |Value| |-------------------|----:| |Avg. |33.03| |IFEval (0-Shot) |78.86| |BBH (3-Shot) |44.17| |MATH Lvl 5 (4-Shot)|13.22| |GPQA (0-shot) |12.98| |MuSR (0-shot) |11.54| |MMLU-PRO (5-shot) |37.43|

This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT3-Gen1-MMMUMAG-gemma-2-9B zelk12/MT3-Gen1-BI-gemma-2-9B The following YAML configuration was used to produce this model: Open LLM Leaderboard Evaluation Results Detailed results can be found here | Metric |Value| |-------------------|----:| |Avg. |31.05| |IFEval (0-Shot) |78.38| |BBH (3-Shot) |44.12| |MATH Lvl 5 (4-Shot)| 3.25| |GPQA (0-shot) |12.86| |MuSR (0-shot) |10.76| |MMLU-PRO (5-shot) |36.96|

GGUF Static: https://huggingface.co/mradermacher/MT-Gen2-MU-gemma-2-MT1RAv0.1t0.25-9B-GGUF This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/recoilme-gemma-2-Ataraxy-9B-v0.1-t0.25 zelk12/MT1-gemma-2-9B The following YAML configuration was used to produce this model:

GGUF Static: https://huggingface.co/mradermacher/MT-Gen2-MM-gemma-2-RAv0.1t0.25MT4G1-9B-GGUF This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: zelk12/MT4-Gen1-gemma-2-9B zelk12/recoilme-gemma-2-Ataraxy-9B-v0.1-t0.25 The following YAML configuration was used to produce this model:

GGUF Static: https://huggingface.co/mradermacher/MT1-Gen3-GP-gemma-2-S4S5-9B-GGUF This is a merge of pre-trained language models created using mergekit. This model was merged using the SLERP merge method. The following models were included in the merge: allknowingroger/Gemmaslerp4-10B allknowingroger/GemmaSlerp5-10B The following YAML configuration was used to produce this model: