Embeddings vs LLMs ...</p>\n<p><a href=\"https://cdn-uploads.huggingface.co/production/uploads/5f1eb362eec0ad2a071ad6e2/-i-PENZRgNBQ7aRHczRQb.png\" rel=\"nofollow\"><img src=\"https://cdn-uploads.huggingface.co/production/uploads/5f1eb362eec0ad2a071ad6e2/-i-PENZRgNBQ7aRHczRQb.png\" alt=\"Screenshot 2026-08-21 at 10.46.05 AM\"></a></p>\n","updatedAt":"2026-08-21T17:46:20.853Z","author":{"_id":"5f1eb362eec0ad2a071ad6e2","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5f1eb362eec0ad2a071ad6e2/nDiBXdLrOTw67lJp_y_WA.jpeg","fullname":"Niklas Muennighoff","name":"Muennighoff","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":185,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.43052414059638977},"editors":["Muennighoff"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/5f1eb362eec0ad2a071ad6e2/nDiBXdLrOTw67lJp_y_WA.jpeg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2608.12875","authors":[{"_id":"6a888e7f3d26296ea3091371","name":"Adnan El Assadi","hidden":false},{"_id":"6a888e7f3d26296ea3091372","name":"Niklas Muennighoff","hidden":false},{"_id":"6a888e7f3d26296ea3091373","name":"Jinhyuk Lee","hidden":false}],"publishedAt":"2026-08-13T00:00:00.000Z","submittedOnDailyAt":"2026-08-21T00:00:00.000Z","title":"The Embedder's Dilemma: LLMs Are Better, but at What Cost?","submittedOnDailyBy":{"_id":"5f1eb362eec0ad2a071ad6e2","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5f1eb362eec0ad2a071ad6e2/nDiBXdLrOTw67lJp_y_WA.jpeg","isPro":false,"fullname":"Niklas Muennighoff","user":"Muennighoff","type":"user","name":"Muennighoff"},"summary":"Should you replace your text-embedding pipeline with a large language model? We answer this with a controlled, cost-aware comparison of ten LLMs across six families and 26 embedding models (118M to 14B parameters) on 37 tasks spanning classification, semantic textual similarity (STS), clustering, pair classification, and retrieval. In aggregate the two paradigms are effectively tied: the best LLM (Gemini 3.1 Pro, 77.6) and the best embedding model (77.2) differ by 0.4 points. Their strengths differ by task: LLMs lead on reasoning-heavy retrieval, embedding models lead on classification, and the two match on clustering, STS, and pair classification. Reaching that parity is expensive. An LLM costs up to 1,431x more than an embedding model of comparable quality (USD 154 vs. USD 0.11 per benchmark pass), and the open LLMs tested process tokens 2.5 to 736x more slowly on the same GPU. Reasoning tokens account for 28 to 81% of LLM inference cost; lower reasoning budgets preserve or improve retrieval quality for most models in our ablation. The Pareto frontier contains the leading embedding models and one LLM, Gemini 3.1 Pro. These results support a division of labour: use embedding models for similarity, classification, and clustering, and reserve LLMs for reasoning-intensive retrieval. Our code, datasets, and results are publicly available at https://github.com/embeddings-benchmark/embedders-dilemma.","upvotes":1,"discussionId":"6a888e7f3d26296ea3091374","ai_summary":"Large language models and dedicated embedding models achieve nearly identical aggregate performance across diverse tasks, but embedding models are far cheaper and faster, supporting a division of labor by task type.","ai_keywords":["text-embedding","large language models","semantic textual similarity","clustering","retrieval","reasoning tokens","Pareto frontier","embedding models"],"ai_summary_model":"thinkingmachines/Inkling-Small"},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"63ac5701c21e60a3e9b58aa7","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/63ac5701c21e60a3e9b58aa7/g6EX7diOpuA94R2ab-rZC.png","isPro":true,"fullname":"Dipankar Sarkar","user":"dipankarsarkar","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2608/2608.12875.md","query":{}}">
The Embedder's Dilemma: LLMs Are Better, but at What Cost?
Abstract
Large language models and dedicated embedding models achieve nearly identical aggregate performance across diverse tasks, but embedding models are far cheaper and faster, supporting a division of labor by task type.
Should you replace your text-embedding pipeline with a large language model? We answer this with a controlled, cost-aware comparison of ten LLMs across six families and 26 embedding models (118M to 14B parameters) on 37 tasks spanning classification, semantic textual similarity (STS), clustering, pair classification, and retrieval. In aggregate the two paradigms are effectively tied: the best LLM (Gemini 3.1 Pro, 77.6) and the best embedding model (77.2) differ by 0.4 points. Their strengths differ by task: LLMs lead on reasoning-heavy retrieval, embedding models lead on classification, and the two match on clustering, STS, and pair classification. Reaching that parity is expensive. An LLM costs up to 1,431x more than an embedding model of comparable quality (USD 154 vs. USD 0.11 per benchmark pass), and the open LLMs tested process tokens 2.5 to 736x more slowly on the same GPU. Reasoning tokens account for 28 to 81% of LLM inference cost; lower reasoning budgets preserve or improve retrieval quality for most models in our ablation. The Pareto frontier contains the leading embedding models and one LLM, Gemini 3.1 Pro. These results support a division of labour: use embedding models for similarity, classification, and clustering, and reserve LLMs for reasoning-intensive retrieval. Our code, datasets, and results are publicly available at https://github.com/embeddings-benchmark/embedders-dilemma.
Community
Embeddings vs LLMs ...
Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.
Tap or paste here to upload images
Cite arxiv.org/abs/2608.12875 in a model README.md to link it from this page.
Cite arxiv.org/abs/2608.12875 in a dataset README.md to link it from this page.
Cite arxiv.org/abs/2608.12875 in a Space README.md to link it from this page.
Discussion (0)
Sign in to join the discussion. Free account, 30 seconds — email code or GitHub.
Sign in →No comments yet. Sign in and be the first to say something.