Public log
Every change, logged
General LLM: Gemini 2.5 Pro takes the crown
First holder recorded for this category and slice.General LLM: GLM 5 takes the crown
First holder recorded for this category and slice.General LLM: Kimi K3 takes the crown
Crown change confirmed by two consecutive snapshots or two independent primary sources.Reasoning: Gemini 1.5 Pro 002 takes the crown
First holder recorded for this category and slice.Reasoning: Gemini 2.5 Pro Exp 03.25 and O3 and O4 Mini and Gemini 2.5 Pro Preview 05.06 and Gemini 2.5 Pro Preview 06.05 are co-SOTA
Crown change confirmed by two consecutive snapshots or two independent primary sources.Reasoning: O3 and Gemini 2.5 Pro Preview 06.05 and GPT-5 and GPT-5 Mini are co-SOTA
Crown change confirmed by two consecutive snapshots or two independent primary sources.Reasoning: GPT-5 Pro and Gemini 3 Pro Preview are co-SOTA
Crown change confirmed by two consecutive snapshots or two independent primary sources.Reasoning: GPT-5 Pro and Gemini 3 Pro Preview and Gemini 2.5 Pro are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: GPT-5 Pro and Gemini 3 Pro Preview and Claude Opus 4.1 are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: GPT-5 Pro and Gemini 3 Pro Preview and Claude Opus 4.6 and Claude Opus 4.1 are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: GPT-5 Pro and Gemini 3 Pro Preview and Claude Opus 4.6 and GLM 5 are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: GPT-5 Pro and Gemini 3 Pro Preview and Claude Opus 4.6 and Claude Sonnet 4.5 are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: Gemini 3.1 Pro Preview and Claude Sonnet 4.5 are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: Gemini 3.1 Pro Preview and GPT-5.4 Pro and Claude Sonnet 4.5 are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: Gemini 3.1 Pro Preview and GPT-5.4 Pro and GPT-5.4 Mini are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: Gemini 3.1 Pro Preview and GPT-5.4 Pro and Muse Spark and GPT-5.4 Mini are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: Gemini 3.1 Pro Preview and GPT-5.4 Pro and Muse Spark and Claude Opus 4.5 are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: Gemini 3.1 Pro Preview and GPT-5.4 Pro and Muse Spark and GLM 5.1 are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: Gemini 3.1 Pro Preview and GPT-5.4 Pro and Muse Spark and GPT-5.5 are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: Gemini 3.1 Pro Preview and GPT-5.4 Pro and Muse Spark and Claude Sonnet 4.6 and GPT-5.4 are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: Gemini 3.1 Pro Preview and GPT-5.4 Pro and Muse Spark and Gemini 3 Pro are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: Gemini 3.1 Pro Preview and GPT-5.4 Pro and Muse Spark and Kimi K3 are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: Gemini 3.1 Pro Preview and GPT-5.4 Pro and Muse Spark and GLM 5.3 are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: Gemini 3.1 Pro Preview and GPT-5.4 Pro and Muse Spark and Claude Fable 5.1 and Gemini 3.7 Flash are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: Gemini 3.1 Pro Preview and GPT-5.4 Pro and Muse Spark and Claude Fable 5.1 and Gemini 3.8 Flash and Gemini 3.7 Flash are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: GPT-6 Astra and Gemini 3.7 Flash are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: GPT-6 Astra and Muse Spark 1.1 are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: GPT-6 Astra and Claude Fable 5 are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: GPT-6 Astra and Claude Opus 5 are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: GPT-6 Astra and Claude Opus 5.5 and Gemini 4 Argon are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: GLM 4.5 takes the crown
First holder recorded for this category and slice.Reasoning: Kimi K2 5 and GLM 5 are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: Kimi K2 5 and GLM 5.1 are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: Kimi K2 5 and Kimi K3 are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: Kimi K2 5 and GLM 5.3 are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: Kimi K2 5 and Kimi K3 are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Math: Gemini 1.5 Pro 002 takes the crown
First holder recorded for this category and slice.Math: Gemini 2.5 Pro Exp 03.25 and O3 and O4 Mini and Gemini 2.5 Pro Preview 05.06 and Gemini 2.5 Pro Preview 06.05 are co-SOTA
Crown change confirmed by two consecutive snapshots or two independent primary sources.Math: O3 and Gemini 2.5 Pro Preview 06.05 and GPT-5 and GPT-5 Mini are co-SOTA
Crown change confirmed by two consecutive snapshots or two independent primary sources.Math: GPT-5 Pro and Gemini 3 Pro Preview are co-SOTA
Crown change confirmed by two consecutive snapshots or two independent primary sources.Math: Gemini 3.1 Pro Preview takes the crown
Crown change confirmed by two consecutive snapshots or two independent primary sources.Math: Gemini 3.1 Pro Preview and GPT-5.4 Pro and Muse Spark are co-SOTA
Crown change confirmed by two consecutive snapshots or two independent primary sources.Math: GLM 4.5 takes the crown
First holder recorded for this category and slice.Coding: Claude Opus 5.5 and Claude Sonnet 5.5 are co-SOTA
First holder recorded for this category and slice.Coding: GLM 5.3 and GLM 5.3 Flash are co-SOTA
First holder recorded for this category and slice.Coding › Agentic: Claude Opus 5.5 and Claude Sonnet 5.5 are co-SOTA
First holder recorded for this category and slice.Coding › Agentic: GLM 5.3 and GLM 5.3 Flash are co-SOTA
First holder recorded for this category and slice.Coding › Web dev: Claude Sonnet 4.5 takes the crown
First holder recorded for this category and slice.Coding › Web dev: GPT-5.5 takes the crown
Crown change confirmed by two consecutive snapshots or two independent primary sources.Coding › Web dev: Minimax M2 5 takes the crown
First holder recorded for this category and slice.Long context: O1 takes the crown
First holder recorded for this category and slice.Long context: ChatGPT 4O 01.29 2025 takes the crown
Crown change confirmed by two consecutive snapshots or two independent primary sources.Long context: ChatGPT 4O 01.29 2025 and Gemini 2.0 Flash 001 are co-SOTA
Crown change confirmed by two consecutive snapshots or two independent primary sources.Long context: Gemini 2.5 Pro Exp 03.25 takes the crown
Crown change confirmed by two consecutive snapshots or two independent primary sources.Long context: O3 takes the crown
Crown change confirmed by two consecutive snapshots or two independent primary sources.Long context: O3 and Grok 4 0709 and GPT-5 are co-SOTA
Crown change confirmed by two consecutive snapshots or two independent primary sources.Long context: DeepSeek R1 takes the crown
First holder recorded for this category and slice.Long context: Minimax M1 takes the crown
Crown change confirmed by two consecutive snapshots or two independent primary sources.Vision: Nova 2.0 Pro Preview takes the crown
First holder recorded for this category and slice.Vision: Claude Sonnet 4.6 and Gemini 3.1 Pro Preview and Claude Opus 4.7 are co-SOTA
Crown change confirmed by two consecutive snapshots or two independent primary sources.Vision: Claude Fable 5 and GPT-5.6 Sol are co-SOTA
Crown change confirmed by two consecutive snapshots or two independent primary sources.Vision: GPT-5.6 Sol and GPT-6 Astra are co-SOTA
Crown change confirmed by two consecutive snapshots or two independent primary sources.Vision: Kimi K2 6 takes the crown
First holder recorded for this category and slice.Vision: Kimi K3 takes the crown
Crown change confirmed by two consecutive snapshots or two independent primary sources.Image generation: Mai Image 2.5 takes the crown
First holder recorded for this category and slice.Image edit: GPT Image 2.5 Sunburst takes the crown
First holder recorded for this category and slice.Video generation: Veo 3 takes the crown
First holder recorded for this category and slice.Factuality: Muse Spark 1.1 takes the crown
First holder recorded for this category and slice.Embeddings: Sentence Transformers Paraphrase Multilingual Mpnet Base V2 and Sentence Transformers Labse and Sentence Transformers Paraphrase Multilingual Minilm L12 V2 are co-SOTA
First holder recorded for this category and slice.Embeddings: Qwen Qwen3 Embedding 8B and Qwen Qwen3 Embedding 4B and Google Gemini Embedding 001 and Nvidia Llama Embed Nemotron 8B and Tencent Kalm Embedding Gemma3 12B 2511 and Bytedance Seed1 6 Embedding 1215 are co-SOTA
Crown change confirmed by two consecutive snapshots or two independent primary sources.Embeddings: Qwen Qwen3 Embedding 8B and Tencent Kalm Embedding Gemma3 12B 2511 and Microsoft Harrier Oss V1 27B are co-SOTA
Crown change confirmed by two consecutive snapshots or two independent primary sources.Embeddings: Sentence Transformers Paraphrase Multilingual Mpnet Base V2 and Sentence Transformers Labse and Sentence Transformers Paraphrase Multilingual Minilm L12 V2 are co-SOTA
First holder recorded for this category and slice.Embeddings: Qwen Qwen3 Embedding 8B and Qwen Qwen3 Embedding 4B and Nvidia Llama Embed Nemotron 8B and Tencent Kalm Embedding Gemma3 12B 2511 are co-SOTA
Crown change confirmed by two consecutive snapshots or two independent primary sources.Embeddings: Qwen Qwen3 Embedding 8B and Tencent Kalm Embedding Gemma3 12B 2511 and Microsoft Harrier Oss V1 27B are co-SOTA
Crown change confirmed by two consecutive snapshots or two independent primary sources.Embeddings: Sentence Transformers Paraphrase Multilingual Mpnet Base V2 and Sentence Transformers Labse and Sentence Transformers Paraphrase Multilingual Minilm L12 V2 are co-SOTA
First holder recorded for this category and slice.Embeddings: Qwen Qwen3 Embedding 8B and Qwen Qwen3 Embedding 4B and Nvidia Llama Embed Nemotron 8B and Tencent Kalm Embedding Gemma3 12B 2511 are co-SOTA
Crown change confirmed by two consecutive snapshots or two independent primary sources.Embeddings: Qwen Qwen3 Embedding 8B and Tencent Kalm Embedding Gemma3 12B 2511 and Microsoft Harrier Oss V1 27B are co-SOTA
Crown change confirmed by two consecutive snapshots or two independent primary sources.
First holder recorded for this category and slice.General LLM: GLM 5 takes the crown
First holder recorded for this category and slice.General LLM: Kimi K3 takes the crown
Crown change confirmed by two consecutive snapshots or two independent primary sources.Reasoning: Gemini 1.5 Pro 002 takes the crown
First holder recorded for this category and slice.Reasoning: Gemini 2.5 Pro Exp 03.25 and O3 and O4 Mini and Gemini 2.5 Pro Preview 05.06 and Gemini 2.5 Pro Preview 06.05 are co-SOTA
Crown change confirmed by two consecutive snapshots or two independent primary sources.Reasoning: O3 and Gemini 2.5 Pro Preview 06.05 and GPT-5 and GPT-5 Mini are co-SOTA
Crown change confirmed by two consecutive snapshots or two independent primary sources.Reasoning: GPT-5 Pro and Gemini 3 Pro Preview are co-SOTA
Crown change confirmed by two consecutive snapshots or two independent primary sources.Reasoning: GPT-5 Pro and Gemini 3 Pro Preview and Gemini 2.5 Pro are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: GPT-5 Pro and Gemini 3 Pro Preview and Claude Opus 4.1 are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: GPT-5 Pro and Gemini 3 Pro Preview and Claude Opus 4.6 and Claude Opus 4.1 are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: GPT-5 Pro and Gemini 3 Pro Preview and Claude Opus 4.6 and GLM 5 are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: GPT-5 Pro and Gemini 3 Pro Preview and Claude Opus 4.6 and Claude Sonnet 4.5 are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: Gemini 3.1 Pro Preview and Claude Sonnet 4.5 are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: Gemini 3.1 Pro Preview and GPT-5.4 Pro and Claude Sonnet 4.5 are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: Gemini 3.1 Pro Preview and GPT-5.4 Pro and GPT-5.4 Mini are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: Gemini 3.1 Pro Preview and GPT-5.4 Pro and Muse Spark and GPT-5.4 Mini are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: Gemini 3.1 Pro Preview and GPT-5.4 Pro and Muse Spark and Claude Opus 4.5 are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: Gemini 3.1 Pro Preview and GPT-5.4 Pro and Muse Spark and GLM 5.1 are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: Gemini 3.1 Pro Preview and GPT-5.4 Pro and Muse Spark and GPT-5.5 are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: Gemini 3.1 Pro Preview and GPT-5.4 Pro and Muse Spark and Claude Sonnet 4.6 and GPT-5.4 are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: Gemini 3.1 Pro Preview and GPT-5.4 Pro and Muse Spark and Gemini 3 Pro are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: Gemini 3.1 Pro Preview and GPT-5.4 Pro and Muse Spark and Kimi K3 are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: Gemini 3.1 Pro Preview and GPT-5.4 Pro and Muse Spark and GLM 5.3 are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: Gemini 3.1 Pro Preview and GPT-5.4 Pro and Muse Spark and Claude Fable 5.1 and Gemini 3.7 Flash are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: Gemini 3.1 Pro Preview and GPT-5.4 Pro and Muse Spark and Claude Fable 5.1 and Gemini 3.8 Flash and Gemini 3.7 Flash are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: GPT-6 Astra and Gemini 3.7 Flash are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: GPT-6 Astra and Muse Spark 1.1 are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: GPT-6 Astra and Claude Fable 5 are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: GPT-6 Astra and Claude Opus 5 are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: GPT-6 Astra and Claude Opus 5.5 and Gemini 4 Argon are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: GLM 4.5 takes the crown
First holder recorded for this category and slice.Reasoning: Kimi K2 5 and GLM 5 are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: Kimi K2 5 and GLM 5.1 are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: Kimi K2 5 and Kimi K3 are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: Kimi K2 5 and GLM 5.3 are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Reasoning: Kimi K2 5 and Kimi K3 are co-SOTA
Primary sources disagree beyond the margin of error, so the crown is shared with low confidence.Math: Gemini 1.5 Pro 002 takes the crown
First holder recorded for this category and slice.Math: Gemini 2.5 Pro Exp 03.25 and O3 and O4 Mini and Gemini 2.5 Pro Preview 05.06 and Gemini 2.5 Pro Preview 06.05 are co-SOTA
Crown change confirmed by two consecutive snapshots or two independent primary sources.Math: O3 and Gemini 2.5 Pro Preview 06.05 and GPT-5 and GPT-5 Mini are co-SOTA
Crown change confirmed by two consecutive snapshots or two independent primary sources.Math: GPT-5 Pro and Gemini 3 Pro Preview are co-SOTA
Crown change confirmed by two consecutive snapshots or two independent primary sources.Math: Gemini 3.1 Pro Preview takes the crown
Crown change confirmed by two consecutive snapshots or two independent primary sources.Math: Gemini 3.1 Pro Preview and GPT-5.4 Pro and Muse Spark are co-SOTA
Crown change confirmed by two consecutive snapshots or two independent primary sources.Math: GLM 4.5 takes the crown
First holder recorded for this category and slice.Coding: Claude Opus 5.5 and Claude Sonnet 5.5 are co-SOTA
First holder recorded for this category and slice.Coding: GLM 5.3 and GLM 5.3 Flash are co-SOTA
First holder recorded for this category and slice.Coding › Agentic: Claude Opus 5.5 and Claude Sonnet 5.5 are co-SOTA
First holder recorded for this category and slice.Coding › Agentic: GLM 5.3 and GLM 5.3 Flash are co-SOTA
First holder recorded for this category and slice.Coding › Web dev: Claude Sonnet 4.5 takes the crown
First holder recorded for this category and slice.Coding › Web dev: GPT-5.5 takes the crown
Crown change confirmed by two consecutive snapshots or two independent primary sources.Coding › Web dev: Minimax M2 5 takes the crown
First holder recorded for this category and slice.Long context: O1 takes the crown
First holder recorded for this category and slice.Long context: ChatGPT 4O 01.29 2025 takes the crown
Crown change confirmed by two consecutive snapshots or two independent primary sources.Long context: ChatGPT 4O 01.29 2025 and Gemini 2.0 Flash 001 are co-SOTA
Crown change confirmed by two consecutive snapshots or two independent primary sources.Long context: Gemini 2.5 Pro Exp 03.25 takes the crown
Crown change confirmed by two consecutive snapshots or two independent primary sources.Long context: O3 takes the crown
Crown change confirmed by two consecutive snapshots or two independent primary sources.Long context: O3 and Grok 4 0709 and GPT-5 are co-SOTA
Crown change confirmed by two consecutive snapshots or two independent primary sources.Long context: DeepSeek R1 takes the crown
First holder recorded for this category and slice.Long context: Minimax M1 takes the crown
Crown change confirmed by two consecutive snapshots or two independent primary sources.Vision: Nova 2.0 Pro Preview takes the crown
First holder recorded for this category and slice.Vision: Claude Sonnet 4.6 and Gemini 3.1 Pro Preview and Claude Opus 4.7 are co-SOTA
Crown change confirmed by two consecutive snapshots or two independent primary sources.Vision: Claude Fable 5 and GPT-5.6 Sol are co-SOTA
Crown change confirmed by two consecutive snapshots or two independent primary sources.Vision: GPT-5.6 Sol and GPT-6 Astra are co-SOTA
Crown change confirmed by two consecutive snapshots or two independent primary sources.Vision: Kimi K2 6 takes the crown
First holder recorded for this category and slice.Vision: Kimi K3 takes the crown
Crown change confirmed by two consecutive snapshots or two independent primary sources.Image generation: Mai Image 2.5 takes the crown
First holder recorded for this category and slice.Image edit: GPT Image 2.5 Sunburst takes the crown
First holder recorded for this category and slice.Video generation: Veo 3 takes the crown
First holder recorded for this category and slice.Factuality: Muse Spark 1.1 takes the crown
First holder recorded for this category and slice.Embeddings: Sentence Transformers Paraphrase Multilingual Mpnet Base V2 and Sentence Transformers Labse and Sentence Transformers Paraphrase Multilingual Minilm L12 V2 are co-SOTA
First holder recorded for this category and slice.Embeddings: Qwen Qwen3 Embedding 8B and Qwen Qwen3 Embedding 4B and Google Gemini Embedding 001 and Nvidia Llama Embed Nemotron 8B and Tencent Kalm Embedding Gemma3 12B 2511 and Bytedance Seed1 6 Embedding 1215 are co-SOTA
Crown change confirmed by two consecutive snapshots or two independent primary sources.Embeddings: Qwen Qwen3 Embedding 8B and Tencent Kalm Embedding Gemma3 12B 2511 and Microsoft Harrier Oss V1 27B are co-SOTA
Crown change confirmed by two consecutive snapshots or two independent primary sources.Embeddings: Sentence Transformers Paraphrase Multilingual Mpnet Base V2 and Sentence Transformers Labse and Sentence Transformers Paraphrase Multilingual Minilm L12 V2 are co-SOTA
First holder recorded for this category and slice.Embeddings: Qwen Qwen3 Embedding 8B and Qwen Qwen3 Embedding 4B and Nvidia Llama Embed Nemotron 8B and Tencent Kalm Embedding Gemma3 12B 2511 are co-SOTA
Crown change confirmed by two consecutive snapshots or two independent primary sources.Embeddings: Qwen Qwen3 Embedding 8B and Tencent Kalm Embedding Gemma3 12B 2511 and Microsoft Harrier Oss V1 27B are co-SOTA
Crown change confirmed by two consecutive snapshots or two independent primary sources.Embeddings: Sentence Transformers Paraphrase Multilingual Mpnet Base V2 and Sentence Transformers Labse and Sentence Transformers Paraphrase Multilingual Minilm L12 V2 are co-SOTA
First holder recorded for this category and slice.Embeddings: Qwen Qwen3 Embedding 8B and Qwen Qwen3 Embedding 4B and Nvidia Llama Embed Nemotron 8B and Tencent Kalm Embedding Gemma3 12B 2511 are co-SOTA
Crown change confirmed by two consecutive snapshots or two independent primary sources.Embeddings: Qwen Qwen3 Embedding 8B and Tencent Kalm Embedding Gemma3 12B 2511 and Microsoft Harrier Oss V1 27B are co-SOTA
Crown change confirmed by two consecutive snapshots or two independent primary sources.
Corrections
Submit a source
Paste a public https URL. The site fetches it, hashes the body, and logs accept or reject. A logged source never hand-edits a verdict.