Open LLM Average open-llm-avg Leaderboard
Mean across all 6 open-llm-leaderboard v2 benchmarks - aggregate model quality score. · Metric: Average (higher is better)
| # | Model | Average | Paper |
|---|---|---|---|
| 1 | CombinHorizon/zetasepic-abliteratedV2-Qwen2.5-32B-Inst-BaseMerge-TIES | 46.76 | — |
| 2 | CombinHorizon/huihui-ai-abliterated-Qwen2.5-32B-Inst-BaseMerge-TIES | 45.66 | — |
| 3 | EVA-UNIT-01/EVA-Qwen2.5-72B-v0.2 | 44.22 | — |
| 4 | Aryanne/QwentileSwap | 43.92 | — |
| 5 | Cran-May/tempmotacilla-cinerea-0308 | 43.64 | — |
| 6 | Daemontatox/Llama3.3-70B-CogniLink | 42.77 | — |
| 7 | BenevolenceMessiah/Qwen2.5-72B-2x-Instruct-TIES-v1.0 | 42.27 | — |
| 8 | 1024m/QWEN-14B-B100 | 41.92 | — |
| 9 | CombinHorizon/Josiefied-abliteratedV4-Qwen2.5-14B-Inst-BaseMerge-TIES | 41.77 | — |
| 10 | CombinHorizon/huihui-ai-abliteratedV2-Qwen2.5-14B-Inst-BaseMerge-TIES | 41.47 | — |
| 11 | Baptiste-HUVELLE-10/LeTriomphant2.2_ECE_iLAB | 41.30 | — |
| 12 | CultriX/Qwen2.5-14B-ReasoningMerge | 40.65 | — |
| 13 | Daemontatox/PathFinderAi3.0 | 40.46 | — |
| 14 | Cran-May/merge_model_20250308_2 | 40.26 | — |
| 15 | CultriX/Qwen2.5-14B-HyperMarck-dl | 39.89 | — |
| 16 | CultriX/Qwen2.5-14B-Brocav3 | 39.85 | — |
| 17 | CultriX/Qwen2.5-14B-Brocav6 | 39.84 | — |
| 18 | CultriX/Qwen2.5-14B-Hyperionv3 | 39.76 | — |
| 19 | CultriX/Qwen2.5-14B-Hyperionv5 | 39.72 | — |
| 20 | CultriX/Qwen2.5-14B-Brocav7 | 39.62 | — |
| 21 | EpistemeAI/DeepThinkers-Phi4 | 39.41 | — |
| 22 | Daemontatox/CogitoZ | 39.38 | — |
| 23 | CultriX/Qwen2.5-14B-BrocaV9 | 39.26 | — |
| 24 | CultriX/Qwen2.5-14B-partialmergept1 | 39.11 | — |
| 25 | CultriX/Qwen2.5-14B-Ultimav2 | 38.84 | — |
| 26 | CultriX/Qwen2.5-14B-MegaMerge-pt2 | 38.80 | — |
| 27 | CultriX/Qwen2.5-14B-MergeStock | 38.74 | — |
| 28 | CultriX/SeQwence-14Bv2 | 38.74 | — |
| 29 | CultriX/SeQwence-14Bv3 | 38.67 | — |
| 30 | CultriX/Qwen2.5-14B-Emergedv3 | 38.66 | — |