Up to date
Models › Qwen3-Embedding family
Qwen Apache 2.0 out 17 months ago
596M parameters, text in, embeddings out, 32K tokens of context. 1 provider serves it from $0.010 per million input tokens.
Verified 2026-10-25 Rank #7 in Embeddings · score 108.2 method v1 Sources 3 Downloads 30d 8.2M · #16 of 619 tracked
Changed this week: 3 new quantized builds on the Hub (272 in all).2026-10-24 · 2 new quantized builds on the Hub (271 in all).2026-10-24
Picked by rule from 7 tracked pieces: the lab's own post, the most shared, a field report, the latest.
The Qwen3 Embedding series introduces new text embedding and reranking models built on Qwen3 foundation models.
The 1 pick above, plus these 6. Top is the radar's score: discussion, source authority and the lab's own posts, fading with age.
Live hosted endpoints, cheapest first, USD per million tokens. Ties break on output price, then provider name.
| # | Cheapest live row | In /1M | Out /1M | Context | Quant | Seen |
|---|---|---|---|---|---|---|
| 1 | DeepInfra | $0.010 | — | 32K | — | Oct 25 |
Hugging Face does not say who downloads a model or with what. These are the proxies: API providers listing it, quantized builds, fine-tunes/adapters/merges, and Spaces that load it.
Served by 1 API provider; heavily run locally (272 quantized builds); a big base for builders (604 fine-tunes, adapters and merges).
Top Spaces running it: embedl/hfviewer ♥ 181 · hugging-apps/mulacover-demo ♥ 44 · hfmlsoc/smollm3-eu-data-transparency ♥ 8 · firefighter/TransDis-CreativityAutoAssessment ♥ 7
Weights only, an estimate: leave room for the KV cache. Method
About 0.4 GB at INT4, 0.7 GB at FP8, 1.4 GB at BF16: weights only, so leave room for the KV cache.
Estimate params × bytes per param × 1.2 · how we estimate
Third-party measurements only; author-reported numbers stay in Reference, marked as such.
70.47. Each tick is a model on this site with a measured MTEB row; author-reported numbers stay in Reference. Mean(task) x100 over 41 tasks. 41 tasks, split=test
Source · leaderboard · 2026-10-24
Weights, every dated article, and repo commits on one rail.
Teal is the weights (2025-06-03), amber triangles are 5 dated articles (each links to it), grey ticks are repo commits.
876 derivatives on the Hub. Which build to run, by runtime: the most downloaded in each class.
From the Hugging Face model tree; quantized builds in teal, fine-tunes, merges and adapters in greys.
Everything else, folded: the full tables, every source.
| Benchmark | Value | Setting / notes | Reported by | Date |
|---|---|---|---|---|
| measured MTEB · Embed text for search | 61.83 nDCG@10 x100 (mean of 10 tasks) | 10 tasks, split=test | leaderboard | 2026-10-24 |
| measured MTEB · Embed text for search | 70.47 mean(task) x100 over 41 tasks | 41 tasks, split=test | leaderboard | 2026-10-24 |
| measured MTEB Reranking · Rerank search results | 48.18 MAP x100 (mean of 2 tasks) | 2 tasks, split=test | leaderboard | 2026-10-24 |
The model is designed for tasks such as text retrieval and classification, supporting over 100 languages.[1] It features a 32k context length and is released under the Apache 2.0 license.[2][3]
Generated 2026-10-07 15:00 UTC by parasail-gemma-4-26b-a4b-it; each numeral points at the fact it rests on. Not a verdict.
Labels are evidence, not verdicts: measured a third-party leaderboard row for this model; author-reported a number the model author published in the card or eval files; structural a deterministic signal from the name, config or chat template, or a decision-model estimate. Each chip links to its source. All use cases.
Listed in Embeddings
| Pipeline tag | feature-extraction |
|---|---|
| Library | sentence-transformers |
| Architectures | Qwen3ForCausalLM |
| Model type | qwen3 |
| Dtype | bfloat16 |
| Params by dtype | BF16: 595,776,512 |
| Size band | S |
| MoE | no |
| Context length | 32,768 (config.json max_position_embeddings) |
| Papers | arXiv:2506.05176 |
| License | Apache 2.0 (apache-2.0) |
| Hub snapshot | 2026-10-25 |
vLLM: supported · Qwen3ForCausalLM · 7050ef881ff0835f6890a47db1859a24c65481d6SGLang: supported · Qwen3ForCausalLM · 773cb938fe9484a8f7e7e46fb1b1fe96412c59ed
vLLM registry (7050ef881ff0835f6890a47db1859a24c65481d6) · SGLang registry (773cb938fe9484a8f7e7e46fb1b1fe96412c59ed) · Matched by architecture class name against each project's registry on main; 'supported' means the architecture is registered, not that this checkpoint was tested.
Weights-only memory for Qwen3-Embedding-0.6B: params × bytes/param × 1.2 overhead. KV cache and activations are extra. How this is computed.
| BF16 (2 bytes/param) | 1.4 GB · Edge (<4 GB) |
|---|---|
| FP8 (1 byte/param) | 0.7 GB · Edge (<4 GB) |
| INT4 (0.5 bytes/param) | 0.4 GB · Edge (<4 GB) |
Weights only (params × bytes/param × 1.2 overhead); KV cache and activations extra.
Base model: Qwen/Qwen3-0.6B-Base
finetune: 571quantized: 272adapter: 32merge: 1
| Variant | Relation | Quant | Params | Band | License | Author | Likes | Downloads |
|---|---|---|---|---|---|---|---|---|
| farbodtavakkoli/OTel-Embedding-600M | finetune | — | 596M | S | apache-2.0 | farbodtavakkoli | 2 | 425.5k |
| mlx-community/Qwen3-Embedding-0.6B-4bit-DWQ | quantized | -4bit | 596M | S | apache-2.0 | mlx-community | 15 | 308.6k |
| Octen/Octen-Embedding-0.6B | finetune | — | 596M | S | apache-2.0 | Octen | 55 | 205.1k |
| onnx-community/Qwen3-Embedding-0.6B-ONNX | quantized | — | — | — | — | onnx-community | 44 | 153.4k |
| binedge/Qwen3-Embedding-0.6B-FP8 | quantized | -FP8 | 596M | S | apache-2.0 | binedge | 1 | 143.3k |
| dolfsai/Qwen3-Embedding-0.6B-vllm-W8A8 | quantized | -W8A8 | 751M | S | apache-2.0 | dolfsai | 1 | 62.6k |
| mhiveai/Qwen-Insure | finetune | — | 596M | S | — | mhiveai | 0 | 41.1k |
| pipizhao/SkillRouter-Embedding-0.6B | finetune | — | 596M | S | apache-2.0 | pipizhao | 8 | 37.6k |
| mlx-community/Qwen3-Embedding-0.6B-8bit | finetune | -8bit | 596M | S | apache-2.0 | mlx-community | 2 | 33.4k |
| PeterAM4/Qwen3-Embedding-0.6B-GGUF | quantized | GGUF | — | — | apache-2.0 | PeterAM4 | 3 | 33k |
| drawais/Qwen3-Embedding-0.6B-AWQ-INT4 | quantized | AWQ | 599M | S | apache-2.0 | drawais | 4 | 27.5k |
| OpenMed/OpenMed-PII-QwenMed-XLarge-600M-v1 | finetune | — | 596M | S | apache-2.0 | OpenMed | 1 | 20.3k |
| OpenVINO/Qwen3-Embedding-0.6B-int8-ov | finetune | int8 | — | — | apache-2.0 | OpenVINO | 1 | 15k |
| n24q02m/Qwen3-Embedding-0.6B-ONNX | quantized | — | — | — | apache-2.0 | n24q02m | 1 | 13.6k |
| kwisschen/qwen3-embedding-0.6b-patent-oa-rejections | adapter | — | 596M | S | cc-by-nc-4.0 | kwisschen | 1 | 12.4k |
| ThakiCloud/SKILLRET-Embedding-0.6B | finetune | — | 596M | S | apache-2.0 | ThakiCloud | 10 | 11.9k |
| enacimie/Qwen3-Embedding-0.6B-Q4_K_M-GGUF | quantized | GGUF | — | — | apache-2.0 | enacimie | 10 | 11.3k |
| mradermacher/Qwen3-Embedding-0.6B-i1-GGUF | quantized | GGUF | — | — | apache-2.0 | mradermacher | 2 | 10.5k |
| xthor/Qwen3-Embedding-0.6B-GraphQL | quantized | GGUF | 596M | S | apache-2.0 | xthor | 6 | 8.7k |
| furiosa-ai/Qwen3-Embedding-0.6B | finetune | — | 596M | S | apache-2.0 | furiosa-ai | 1 | 8.8k |
| mradermacher/Qwen3-Embedding-0.6B-GGUF | quantized | GGUF | — | — | apache-2.0 | mradermacher | 2 | 7.9k |
| prestoai/qwen3-embedding-0.6b-arabic-ecom | adapter | — | 596M | S | apache-2.0 | prestoai | 1 | 7k |
| boboliu/Qwen3-Embedding-0.6B-W4A16-G128 | quantized | -W4A16 | 599M | S | apache-2.0 | boboliu | 4 | 6.7k |
| Day1Kim/Qwen3-Embedding-0.6B-Korean | finetune | — | 596M | S | — | Day1Kim | 4 | 6k |
| yomir/Qwen3-Embedding-0.6B-GGUF | quantized | GGUF | — | — | apache-2.0 | yomir | 0 | 5.8k |
| futur/Qwen3-Embedding-0.6B-model2vec-onnx | quantized | — | 38.8M | S | mit | futur | 4 | 5.4k |
| Omartificial-Intelligence-Space/Semantic-Ar-Qwen-Embed-0.6B | finetune | — | 596M | S | apache-2.0 | Omartificial-Intelligence-Space | 8 | 4.8k |
| kerncore/Qwen3-Embedding-0.6B-MXL-4bit | quantized | -4bit | 596M | S | apache-2.0 | kerncore | 2 | 4.9k |
| tencent/R3-embedding-0.6b | finetune | — | 596M | S | apache-2.0 | tencent | 27 | 3.5k |
| shawnw3i/Qwen3-Embedding-0.6B-ONNX | quantized | — | — | — | apache-2.0 | shawnw3i | 7 | 4.5k |
| Qdrant/Qwen3-Embedding-0.6B-onnx | quantized | — | — | — | apache-2.0 | Qdrant | 2 | 4.6k |
| Hanno-Labs/dinghy-law-0.6b-v1 | finetune | — | 596M | S | apache-2.0 | Hanno-Labs | 3 | 4.1k |
| JusteLeo/Qwen3-0.6B-T5-xxl-GGUF | quantized | GGUF | — | — | apache-2.0 | JusteLeo | 2 | 4k |
| cstr/qwen3-embed-0.6b-GGUF | quantized | GGUF | — | — | apache-2.0 | cstr | 0 | 4.1k |
| AQ-MedAI/Diver-Retriever-0.6B | finetune | — | 596M | S | apache-2.0 | AQ-MedAI | 7 | 3.6k |
| Casual-Autopsy/Qwen3-Embedding-0.6B-GGUFs | quantized | GGUF | — | — | apache-2.0 | Casual-Autopsy | 0 | 3.7k |
| Gideon531/Qwen3-Embedding-0.6B-Q8_0-GGUF | quantized | Q8_0 | — | — | apache-2.0 | Gideon531 | 4 | 2.9k |
| zenlm/zen-embedding-0.6B-GGUF | quantized | GGUF | — | — | apache-2.0 | zenlm | 2 | 2.9k |
| Svenni551/Qwen3-Embedding-0.6B-ONNX-INT8 | quantized | INT8 | — | — | apache-2.0 | Svenni551 | 1 | 2.8k |
| alphaedge-ai/Qwen3-Embedding-ron-32768 | quantized | — | 474M | S | apache-2.0 | alphaedge-ai | 1 | 2.7k |
| Date | Commit | SHA |
|---|---|---|
| 2026-04-20 | Update README.md (#55) | 97b0c61 |
| 2025-06-20 | Update README.md with TEI support (#26) | c54f2e6 |
| 2025-06-07 | add use case for vllm & modify Citation (#5) | 7441690 |
| 2025-06-06 | Automatically add EOS via Tokenizer, integrate Sentence Transformers (#2) | b22da49 |
| 2025-06-05 | Update README.md | a579a21 |
| 2025-06-05 | Update README.md | 52f1abe |
| 2025-06-05 | Delete special_tokens_map.json | d38d10e |
| 2025-06-05 | Delete added_tokens.json | d43997c |
| 2025-06-05 | Update README.md | e692b5a |
| 2025-06-05 | Update README.md | 68f00aa |
| 2025-06-05 | Update README.md | c935f2d |
| 2025-06-03 | Upload folder using huggingface_hub | 99cabfa |
| 2025-06-03 | Upload folder using huggingface_hub | 1c68a84 |
| 2025-06-03 | initial commit | f2c7d11 |
Hosted endpoints observed for Qwen3-Embedding-0.6B, sorted by input price (USD per million tokens). Each row names its observer and date; prices change daily, so check the source before quoting. Cheapest observed input price: $0.01/M.
| Provider | Model ref | Context | Quant | $/M in | $/M out | tok/s | TTFT | Tools | Status | Observed |
|---|---|---|---|---|---|---|---|---|---|---|
| DeepInfra hf, deepinfra | Qwen/Qwen3-Embedding-0.6B | 32K | — | $0.01 | — | — | — | — | live | 2026-10-25 |
Ranked by source signals (HN, Lobsters, authority domains, official posts) and recency. Summaries are generated from a stored copy of the article text; every sentence cites a span of that text, and nothing is shown that failed the citation check.
| Title | Source | Published | First seen | Signals |
|---|---|---|---|---|
| [PDF] Qwen3 Embedding: Advancing Text Embedding and ... - arXiv | arxiv.org | — | 2026-10-07 | exa |
| Qwen3 embeddings and rerankers: 0.6B vs 4B, text vs VL, and where ColQwen fits | superlinked.com | — | 2026-10-07 | exa |
| Fine-Tuning Qwen3 Embeddings for product category classification | blog.ivan.digital | 2026-03-03 | 2026-10-07 | hn |
| Qwen3 Embedding: Advancing Text Embedding and Reranking with Foundation Models | arxiv.org | 2025-06-11 | 2026-10-07 | hn |
| Qwen3 Embedding:Advancing Text Embedding and Reranking Through Foundation Models | qwenlm.github.io official | 2025-06-05 | 2026-10-07 | hn, exa |
| QwenLM/Qwen3-Embedding | github.com | 2025-06-05 | 2026-10-07 | exa |
| Qwen3 Embedding: Advancing Text Embedding and Reranking Through Foundation Models — Zen Blog | zenlm.org | 2025-06-04 | 2026-10-07 | exa |
arXiv links found in the model card and its tags. "Peer-reviewed" is shown only with evidence: a published venue in OpenAlex (measured) or the authors' own journal reference on arXiv (reported).
Hub snapshot: 2026-10-25 · HF API: 2026-10-24 21:33 UTC · Providers: 2026-10-24 21:44 UTC · Articles: 2026-10-24 21:45 UTC · Summary: 2026-10-07 15:00 UTC · Last verified 2026-10-25
Email alerts are not open yet; the feeds above are the same data. See feeds for the item schema.
