License name consistency (for quantification purposes)

DBlankvoort · DBlankvoort · commit 2201f9934206 · 2025-06-10T16:16:12.000+02:00
diff --git a/AltDiffusion.yaml b/AltDiffusion.yaml
@@ -20,7 +20,7 @@ system:
     performanceclass: limited
     basemodelname: Stable-Diffusion-2-Base
     endmodelname: AltDiffusion
-    endmodellicense: CreativeML Open RAIL-M License
+    endmodellicense: CreativeML Open RAIL-M
     releasedate: 2022-12
     notes: Early multilingual text-to-image model.
 
diff --git a/AnimateDiff-Lightning.yaml b/AnimateDiff-Lightning.yaml
@@ -20,7 +20,7 @@ system:
     performanceclass: limited
     basemodelname: AnimateDiff-SD1.5-v2
     endmodelname: AnimateDiff-Lightning
-    endmodellicense: CreativeML-OpenRAIL-M
+    endmodellicense: CreativeML Open RAIL-M
     releasedate: 2024-03
     notes: Video generation model using progressive adversarial diffusion distillation. Distilled from various models.
 
diff --git a/BELLE.yaml b/BELLE.yaml
@@ -20,7 +20,7 @@ system:
     performanceclass: limited
     basemodelname: Llama-2-13B
     endmodelname: BELLE-Llama2-13B-Chat-0.4M
-    endmodellicense: Apache License 2.0
+    endmodellicense: Apache-2.0
     releasedate: 2023-03
     notes: Llama-2 fine-tuned on a small amount of Chinese data.
 
diff --git a/BERT.yaml b/BERT.yaml
@@ -20,7 +20,7 @@ system:
     performanceclass: limited
     basemodelname: BERT
     endmodelname: BERT
-    endmodellicense: Apache 2.0
+    endmodellicense: Apache-2.0
     releasedate: 2018-11
     notes: "This entry collates base BERT and its many derivatives"
 
diff --git a/Baichuan.yaml b/Baichuan.yaml
@@ -20,7 +20,7 @@ system:
     performanceclass: full
     basemodelname: Baichuan2-13B-Base
     endmodelname: Baichuan2-13B-Chat
-    endmodellicense: Apache-2.0 + Community License for Baichuan2 Model
+    endmodellicense: Apache-2.0 and Community License for Baichuan2 Model
     releasedate: 2023-07
     notes: Bilingual Chinese-English model
 
diff --git a/Claire.yaml b/Claire.yaml
@@ -20,7 +20,7 @@ system:
     performanceclass: full
     basemodelname: Falcon-7B
     endmodelname: Claire-7B-FR-Instruct-0.1
-    endmodellicense: CC-BY-NC-SA 4.0
+    endmodellicense: CC-BY-NC-SA-4.0
     releasedate: 2024-09
     notes: Falcon-based French-language LLM.
 
diff --git a/CodeGemma.yaml b/CodeGemma.yaml
@@ -20,7 +20,7 @@ system:
     performanceclass: full
     basemodelname: Gemma-7B
     endmodelname: CodeGemma-1.1-7B-IT
-    endmodellicense: Gemma license
+    endmodellicense: Gemma Terms of Use
     releasedate: 2024-07
     notes: Gemma-based coder model.
 
diff --git a/CodeGen.yaml b/CodeGen.yaml
@@ -20,7 +20,7 @@ system:
     performanceclass: full
     basemodelname: CodeGen25-7B-Multi-P
     endmodelname: CodeGen25-7B-Instruct-P
-    endmodellicense: unknown (research purposes only)
+    endmodellicense: Unknown (research purposes only)
     releasedate: 2023-06
     notes: Autoregressive language model for code synthesis. Base model licensed under Apache 2.0.
 
diff --git a/CodeLlama.yaml b/CodeLlama.yaml
@@ -20,7 +20,7 @@ system:
     performanceclass: full
     basemodelname: Llama-2-70B
     endmodelname: CodeLlama-70B-Instruct
-    endmodellicense: Llama 2 License
+    endmodellicense: Llama 2 Community License Agreement
     releasedate: 2024-03
     notes: Coder model by Meta.
 
diff --git a/DeepSeek-Coder.yaml b/DeepSeek-Coder.yaml
@@ -20,7 +20,7 @@ system:
     performanceclass: latest
     basemodelname: DeepSeek-V2
     endmodelname: DeepSeek-Coder-V2-Instruct-0724
-    endmodellicense: DeepSeek License
+    endmodellicense: DeepSeek License Agreement
     releasedate: 2024-09
     notes: Continued pretrained from an intermediate checkpoint of DeepSeek-v2; model inheritance is arguable.
 
diff --git a/MPT.yaml b/MPT.yaml
@@ -20,7 +20,7 @@ system:
     performanceclass: full
     basemodelname: MPT-30B
     endmodelname: MPT-30B-Instruct
-    endmodellicense: Apache 2.0
+    endmodellicense: Apache-2.0
     releasedate: 2023-06
     notes: Open LLM by Databricks.
 
diff --git a/Nanbeige.yaml b/Nanbeige.yaml
@@ -20,7 +20,7 @@ system:
     performanceclass: limited
     basemodelname: Nanbeige2-16B
     endmodelname: Nanbeige2-16B-Chat
-    endmodellicense: Apache 2.0 and bespoke community license
+    endmodellicense: Apache-2.0 and bespoke community license
     releasedate: 2024-03
     notes: Comes in 8B and 16B versions
 
diff --git a/Neo.yaml b/Neo.yaml
@@ -20,7 +20,7 @@ system:
     performanceclass: full
     basemodelname: Neo-7B
     endmodelname: Neo-7B-Instruct-v0.1
-    endmodellicense: Apache 2.0
+    endmodellicense: Apache-2.0
     releasedate: 2024-05
     notes: Language model aiming to be completely open-source.
 
diff --git a/NeuralChat.yaml b/NeuralChat.yaml
@@ -20,7 +20,7 @@ system:
     performanceclass: full
     basemodelname: Mistral-7B-v0.1
     endmodelname: Neural-Chat-7B-v3.3
-    endmodellicense: Apache 2.0
+    endmodellicense: Apache-2.0
     releasedate: 2023-11
     notes: Mistral-based Orca-finetuned chat model.
 
diff --git a/OLMo.yaml b/OLMo.yaml
@@ -20,7 +20,7 @@ system:
     performanceclass: full
     basemodelname: OLMo-2-0325-32B
     endmodelname: OLMo-2-0325-32B-Instruct
-    endmodellicense: Apache 2.0
+    endmodellicense: Apache-2.0
     releasedate: 2024-11
     notes: Open LLM trained from scratch by Allen AI.
 
diff --git a/Open-Assistant.yaml b/Open-Assistant.yaml
@@ -20,7 +20,7 @@ system:
     performanceclass: full
     basemodelname: Pythia-12B
     endmodelname: Pythia-12B-SFT-v8-7K-Steps
-    endmodellicense: Apache 2.0
+    endmodellicense: Apache-2.0
     releasedate: 2023-02
     notes: Community-built model aiming to bridge the gap between open-source models and ChatGPT.
 
diff --git a/Poro.yaml b/Poro.yaml
@@ -20,7 +20,7 @@ system:
     performanceclass: full
     basemodelname: Poro-34B
     endmodelname: Poro-34B
-    endmodellicense: Apache 2.0 
+    endmodellicense: Apache-2.0 
     releasedate: 2024-04
     notes: Bilingual model trained on Finnish, English, and programming languages
 
diff --git a/Solar.yaml b/Solar.yaml
@@ -20,7 +20,7 @@ system:
     performanceclass: latest
     basemodelname: Mistral-7B-v0.1
     endmodelname: SOLAR-10.7B-Instruct-v1.0
-    endmodellicense: Meta Community license, CC-BY-NC 4.0
+    endmodellicense: Meta Community License and CC-BY-NC 4.0
     releasedate: 2023-12
     notes: HuggingFace profile says 'Solar is a great example of the progress enabled by open source.'
 
diff --git a/T5.yaml b/T5.yaml
@@ -20,7 +20,7 @@ system:
     performanceclass: limited
     basemodelname: T5
     endmodelname: T5X
-    endmodellicense: Apache 2.0
+    endmodellicense: Apache-2.0
     releasedate: 2019-10
     notes: "This entry collates T5 base, T5X, and FLAN T5"
 
diff --git a/Teuken.yaml b/Teuken.yaml
@@ -20,7 +20,7 @@ system:
     performanceclass: full
     basemodelname: Teuken-7B-base
     endmodelname: Teuken-7B-instruct
-    endmodellicense: Apache 2.0
+    endmodellicense: Apache-2.0
     releasedate: 2024-09
     notes: Open-source multilingual LLM that claims to support all 24 official languages of the European Union.
 
diff --git a/WaveCoder.yaml b/WaveCoder.yaml
@@ -20,7 +20,7 @@ system:
     performanceclass: full
     basemodelname: Deepseek-Coder-6.7B-Base
     endmodelname: WaveCoder-Ultra-6.7B
-    endmodellicense: MIT license
+    endmodellicense: MIT License
     releasedate: 2024-07
     notes: Not absolutely clear that base rather than instruct-tuned model was used, but can be assumed.
 
diff --git a/WizardLM.yaml b/WizardLM.yaml
@@ -20,7 +20,7 @@ system:
     performanceclass: full
     basemodelname: LLaMA-7B
     endmodelname: Evol-Instruct (synthetic)
-    endmodellicense: Llama 2 license
+    endmodellicense: Llama 2 Community License Agreement
     releasedate: 2023-04
     notes: Empowering Large Pre-Trained Language Models to Follow Complex Instructions
 
diff --git a/XGen.yaml b/XGen.yaml
@@ -20,7 +20,7 @@ system:
     performanceclass: limited
     basemodelname: XGen-Small-9B-Base-R 
     endmodelname: XGen-Small-9B-Instruct-R
-    endmodellicense: Research purposes only
+    endmodellicense: CC-BY-NC-4.0
     releasedate: 2025-05
     notes: Compact enterprise-ready LLM.
 
diff --git a/Yi-Coder.yaml b/Yi-Coder.yaml
@@ -20,7 +20,7 @@ system:
     performanceclass: latest
     basemodelname: Yi-9B
     endmodelname: Yi-Coder-9B-Chat
-    endmodellicense: Apache 2.0
+    endmodellicense: Apache-2.0
     releasedate: 2024-07 
     notes: Arguable that Yi-6B is the base model instead, since Yi-9B was created from Yi-6B. Unclear whether the code model was based on the regular 9B model or the context-length-extended one.
 
diff --git a/Yi.yaml b/Yi.yaml
@@ -20,7 +20,7 @@ system:
     performanceclass: full
     basemodelname: Yi-34B
     endmodelname: Yi-1.5-34B-Chat-16K
-    endmodellicense: Apache 2.0
+    endmodellicense: Apache-2.0
     releasedate: 2024-05
     notes: The Yi series models are the next generation of open-source large language models trained from scratch by 01.AI. Targeted as a bilingual language model and trained on 3T multilingual corpus.
 
diff --git a/amber.yaml b/amber.yaml
@@ -20,7 +20,7 @@ system:
     performanceclass: limited
     basemodelname: Amber
     endmodelname: AmberSafe
-    endmodellicense: Apache 2.0
+    endmodellicense: Apache-2.0
     releasedate: 2023-11
     notes: Model released to make LLM training knowledge accessible to all.
 
diff --git a/bloomz.yaml b/bloomz.yaml
@@ -20,7 +20,7 @@ system:
     performanceclass: full
     basemodelname: BLOOM
     endmodelname: BLOOMZ-P3
-    endmodellicense: Apache 2.0 and RAIL (responsible AI license)
+    endmodellicense: Apache-2.0 and RAIL
     releasedate: 2022-09
     notes: Large open multilingual language model.
 
diff --git a/command-r.yaml b/command-r.yaml
@@ -20,7 +20,7 @@ system:
     performanceclass: full
     basemodelname: C4AI-Command-R-V01
     endmodelname: Aya-23-35B
-    endmodellicense: CC-BY-NC and C4AI acceptable use policy
+    endmodellicense: CC-BY-NC-4.0 and C4AI acceptable use policy
     releasedate: 2024-03
     notes: Multilingual open LLM.
 
diff --git a/dolly.yaml b/dolly.yaml
@@ -20,7 +20,7 @@ system:
     performanceclass: full
     basemodelname: Pythia-12B
     endmodelname: Dolly-V2-12B
-    endmodellicense: MIT license
+    endmodellicense: MIT License
     releasedate: 2023-03
     notes: Pythia-based instruction-following model.
 
diff --git a/eurus.yaml b/eurus.yaml
@@ -20,7 +20,7 @@ system:
     performanceclass: limited
     basemodelname: Mixtral-8x22B-v0.1
     endmodelname: Eurux-8x22B-NCA
-    endmodellicense: Apache 2.0 
+    endmodellicense: Apache-2.0 
     releasedate: 2024-04
     notes: Reasoning MoE model.
 
diff --git a/minerva.yaml b/minerva.yaml
@@ -20,7 +20,7 @@ system:
     performanceclass: full
     basemodelname: Minerva-7B-base-v1.0
     endmodelname: Minerva-7B-instruct-v1.0
-    endmodellicense: Apache 2.0
+    endmodellicense: Apache-2.0
     releasedate: 2024-11
     notes: LLM pretrained from scratch on Italian.
 
diff --git a/mistral-nemo.yaml b/mistral-nemo.yaml
@@ -20,7 +20,7 @@ system:
     performanceclass: full
     basemodelname: Mistral-NeMo-12B-Base
     endmodelname: Mistral-NeMo-12B-Instruct
-    endmodellicense: Apache 2.0 (model weights only)
+    endmodellicense: Apache-2.0
     releasedate: 2024-07
     notes: Multilingual model trained by Mistral AI.
 
diff --git a/openjourney.yaml b/openjourney.yaml
@@ -20,7 +20,7 @@ system:
     performanceclass: full
     basemodelname: Stable-Diffusion-v1.5
     endmodelname: OpenJourney-v4
-    endmodellicense: creativeml-openrail-m
+    endmodellicense: CreativeML Open RAIL-M
     releasedate: 2022-11
     notes: Model trained on MidJourney images.
 
diff --git a/viking.yaml b/viking.yaml
@@ -20,7 +20,7 @@ system:
     performanceclass: full
     basemodelname: Viking-33B
     endmodelname: Viking-33B
-    endmodellicense:  Apache 2.0 
+    endmodellicense:  Apache-2.0 
     releasedate: 2024-05
     notes: Multilingual model trained on Nordic languages, English, and code. Also available as 7B and 13B models.
 
