Skip to content

Commit 0b00780

Browse files
xyoniumxyonium
authored
Add tags to qwen3-vl-reranker and Qwen3-VL-Embedding to the gallery (#9628)
* Add tags to Qwen3-VL-Reranker models Added tags for reranker models in index.yaml. Signed-off-by: ER-EPR <38782737+ER-EPR@users.noreply.github.com> * Add Qwen3-VL-Embedding models to gallery Added Qwen3-VL-Embedding-8B and Qwen3-VL-Embedding-2B models with detailed descriptions and file references. Signed-off-by: ER-EPR <38782737+ER-EPR@users.noreply.github.com> * Update index.yaml Signed-off-by: ER-EPR <38782737+ER-EPR@users.noreply.github.com> --------- Signed-off-by: ER-EPR <38782737+ER-EPR@users.noreply.github.com>
1 parent 80961d2 commit 0b00780

1 file changed

Lines changed: 121 additions & 0 deletions

File tree

‎gallery/index.yaml‎

Lines changed: 121 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -2233,10 +2233,129 @@
22332233
- filename: llama-cpp/models/GLM-4.7-Flash-Q4_K_M.gguf
22342234
uri: https://huggingface.co/unsloth/GLM-4.7-Flash-GGUF/resolve/main/GLM-4.7-Flash-Q4_K_M.gguf
22352235
sha256: 29837ed2c0fc5f51981adf8ac8083fcf80743c598381f13e9f06cbad0498b174
2236+
2237+
- name: "qwen3-vl-embedding-8b"
2238+
url: "github:mudler/LocalAI/gallery/virtual.yaml@master"
2239+
urls:
2240+
- https://huggingface.co/VesNFF/Qwen3-VL-Embedding-8B-GGUF
2241+
- https://huggingface.co/Qwen/Qwen3-VL-Embedding-8B
2242+
tags:
2243+
- embedding
2244+
icon: https://cdn-avatars.huggingface.co/v1/production/uploads/620760a26e3b7210c2ff1943/-s1gyJfvbE1RgO5iBeNOi.png
2245+
description: |
2246+
**Model Name:** Qwen3-VL-Embedding-8B
2247+
**Base Model:** Qwen/Qwen3-VL-8B-Instruct
2248+
2249+
**Description:**
2250+
The **Qwen3-VL-Embedding** and **Qwen3-VL-Reranker** model series are the latest additions to the Qwen family, built upon the recently open-sourced and powerful Qwen3-VL foundation model. Specifically designed for multimodal information retrieval and cross-modal understanding, this suite accepts diverse inputs including text, images, screenshots, and videos, as well as inputs containing a mixture of these modalities.
2251+
2252+
**Key Features:**
2253+
- Model Type: MultiModal Embedding
2254+
- Supported Languages: 30+ Languages
2255+
- Supported Input Modalities: Text, images, screenshots, videos, and arbitrary multimodal combinations (e.g., text + image, text + video)
2256+
- Number of Parameters: 8B
2257+
- Context Length: 32k
2258+
- Embedding Dimension: Up to 4096, supports user-defined output dimensions ranging from 64 to 4096
2259+
2260+
**Downloads:**
2261+
- [GGUF Files](https://huggingface.co/Qwen/Qwen3-VL-Embedding-8B) (e.g., `Qwen3-VL-Embedding-8B-Q8_0.gguf`).
2262+
2263+
**Usage:**
2264+
- Requires `transformers`, `qwen-vl-utils`, and `torch`.
2265+
- Example: `from scripts.qwen3_vl_embedding import Qwen3VLEmbedder model = Qwen3VLEmbedder(...)`
2266+
2267+
**Citation:**
2268+
@article{qwen3vlembedding, ...}
2269+
2270+
This description emphasizes its capabilities, efficiency, and versatility for multimodal search tasks.
2271+
overrides:
2272+
embeddings: true
2273+
parameters:
2274+
model: llama-cpp/models/Qwen3-VL-Embedding-8B-Q6_K.gguf
2275+
name: Qwen3-VL-Embedding-8B-GGUF
2276+
backend: llama-cpp
2277+
template:
2278+
use_tokenizer_template: true
2279+
known_usecases:
2280+
- chat
2281+
function:
2282+
grammar:
2283+
disable: true
2284+
mmproj: llama-cpp/mmproj/mmproj-Qwen3-VL-Embedding-8B-f16.gguf
2285+
description: Imported from https://huggingface.co/VesNFF/Qwen3-VL-Embedding-8B-GGUF
2286+
options:
2287+
- use_jinja:true
2288+
files:
2289+
- filename: llama-cpp/models/Qwen3-VL-Embedding-8B-Q6_K.gguf
2290+
sha256: 10ee47c017d73f5df31e41669d9600abdfe80c701c77630504108d56f79b48d7
2291+
uri: https://huggingface.co/VesNFF/Qwen3-VL-Embedding-8B-GGUF/resolve/main/Qwen3-VL-Embedding-8B-Q6_K.gguf
2292+
- filename: llama-cpp/mmproj/mmproj-Qwen3-VL-Embedding-8B-f16.gguf
2293+
sha256: 6f104e4299dfd0738ef1b44f4eecdde9dc049d10a73ce69472e0bfbbd687a034
2294+
uri: https://huggingface.co/VesNFF/Qwen3-VL-Embedding-8B-GGUF/resolve/main/mmproj-Qwen3-VL-Embedding-8B-f16.gguf
2295+
- name: "qwen3-vl-embedding-2b"
2296+
url: "github:mudler/LocalAI/gallery/virtual.yaml@master"
2297+
urls:
2298+
- https://huggingface.co/DevQuasar/Qwen.Qwen3-VL-Embedding-2B-GGUF
2299+
- https://huggingface.co/Qwen/Qwen3-VL-Embedding-2B
2300+
tags:
2301+
- embedding
2302+
icon: https://cdn-avatars.huggingface.co/v1/production/uploads/620760a26e3b7210c2ff1943/-s1gyJfvbE1RgO5iBeNOi.png
2303+
description: |
2304+
**Model Name:** Qwen3-VL-Embedding-2B
2305+
**Base Model:** Qwen/Qwen3-VL-2B-Instruct
2306+
2307+
**Description:**
2308+
The **Qwen3-VL-Embedding** and **Qwen3-VL-Reranker** model series are the latest additions to the Qwen family, built upon the recently open-sourced and powerful Qwen3-VL foundation model. Specifically designed for multimodal information retrieval and cross-modal understanding, this suite accepts diverse inputs including text, images, screenshots, and videos, as well as inputs containing a mixture of these modalities.
2309+
2310+
**Key Features:**
2311+
- Model Type: MultiModal Embedding
2312+
- Supported Languages: 30+ Languages
2313+
- Supported Input Modalities: Text, images, screenshots, videos, and arbitrary multimodal combinations (e.g., text + image, text + video)
2314+
- Number of Parameters: 2B
2315+
- Context Length: 32k
2316+
- Embedding Dimension: Up to 2048, supports user-defined output dimensions ranging from 64 to 2048
2317+
2318+
**Downloads:**
2319+
- [GGUF Files](https://huggingface.co/Qwen/Qwen3-VL-Embedding-2B) (e.g., `Qwen3-VL-Embedding-2B-Q8_0.gguf`).
2320+
2321+
**Usage:**
2322+
- Requires `transformers`, `qwen-vl-utils`, and `torch`.
2323+
- Example: `from scripts.qwen3_vl_embedding import Qwen3VLEmbedder model = Qwen3VLEmbedder(...)`
2324+
2325+
**Citation:**
2326+
@article{qwen3vlembedding, ...}
2327+
2328+
This description emphasizes its capabilities, efficiency, and versatility for multimodal search tasks.
2329+
overrides:
2330+
embeddings: true
2331+
parameters:
2332+
model: llama-cpp/models/Qwen3-VL-Embedding-2B.Q8_0.gguf
2333+
name: Qwen3-VL-Embedding-2B-GGUF
2334+
backend: llama-cpp
2335+
template:
2336+
use_tokenizer_template: true
2337+
known_usecases:
2338+
- chat
2339+
function:
2340+
grammar:
2341+
disable: true
2342+
mmproj: llama-cpp/mmproj/mmproj-Qwen3-VL-Embedding-2B.f16.gguf
2343+
description: Imported from https://huggingface.co/DevQuasar/Qwen.Qwen3-VL-Embedding-2B-GGUF
2344+
options:
2345+
- use_jinja:true
2346+
files:
2347+
- filename: llama-cpp/models/Qwen3-VL-Embedding-2B.Q8_0.gguf
2348+
sha256: 7552c2f699c546ce46abd6b66b2aa16ae667c88c830efbd352b12224d4613492
2349+
uri: https://huggingface.co/DevQuasar/Qwen.Qwen3-VL-Embedding-2B-GGUF/resolve/main/Qwen.Qwen3-VL-Embedding-2B.Q8_0.gguf
2350+
- filename: llama-cpp/mmproj/mmproj-Qwen3-VL-Embedding-2B.f16.gguf
2351+
sha256: 3f89a7768ffa6606935319f71bf56bb71871249ba549bf1080a0caea7a088613
2352+
uri: https://huggingface.co/DevQuasar/Qwen.Qwen3-VL-Embedding-2B-GGUF/resolve/main/mmproj-Qwen.Qwen3-VL-Embedding-2B.f16.gguf
22362353
- name: "qwen3-vl-reranker-8b"
22372354
url: "github:mudler/LocalAI/gallery/virtual.yaml@master"
22382355
urls:
22392356
- https://huggingface.co/mradermacher/Qwen3-VL-Reranker-8B-GGUF
2357+
tags:
2358+
- reranker
22402359
description: |
22412360
**Model Name:** Qwen3-VL-Reranker-8B
22422361
**Base Model:** Qwen/Qwen3-VL-Reranker-8B
@@ -2290,6 +2409,8 @@
22902409
url: "github:mudler/LocalAI/gallery/virtual.yaml@master"
22912410
urls:
22922411
- https://huggingface.co/mradermacher/Qwen3-VL-Reranker-2B-i1-GGUF
2412+
tags:
2413+
- reranker
22932414
description: |
22942415
**Model Name:** Qwen3-VL-Reranker-2B-i1
22952416
**Base Model:** Qwen/Qwen3-VL-Reranker-2B

0 commit comments

Comments
 (0)