Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion ollamaqwen359bv2/Chart.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -3,4 +3,4 @@ appversion: '1.0.1'
description: description
name: ollamaqwen359bv2
type: application
version: '1.0.12'
version: '1.0.18'
11 changes: 9 additions & 2 deletions ollamaqwen359bv2/OlaresManifest.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -8,9 +8,9 @@ metadata:
description: "Open-source multimodal models that delivers exceptional utility and performance"
appid: ollamaqwen359bv2
title: Qwen3.5 9B Q4_K_M (Ollama)
version: '1.0.12'
version: '1.0.18'
categories:
- AI
- AI
sharedEntrances:
- name: ollamaqwen359bv2
host: sharedentrances-api
Expand Down Expand Up @@ -52,6 +52,8 @@ spec:
Next-Generation Training Infrastructure: Near-100% multimodal training efficiency compared to text-only training and asynchronous RL frameworks supporting massive-scale agent scaffolds and environment orchestration.

upgradeDescription: |

Replace the Bitnami OpenResty proxy with official OpenResty and run it as UID/GID 1000 with writable runtime paths.
Upgrade Ollama runtime to ollama/ollama:0.30.10.
Upgrade Olares API proxy to beclab/harveyff-olares-ollama:v0.1.22 (OpenAI vision image_url forwarding).
developer: Alibaba
Expand All @@ -60,6 +62,11 @@ spec:
locale:
- en-US
- zh-CN
- de-DE
- es-ES
- it-IT
- fr-FR
- ja-JP
license:
- text: Apache-2.0
url: https://ollama.com/library/qwen3.5:9b
Expand Down
28 changes: 28 additions & 0 deletions ollamaqwen359bv2/i18n/de-DE/OlaresManifest.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,28 @@
metadata:
title: Qwen3.5 9B Q4_K_M (Ollama)
description: "Open-Source-multimodale Modelle mit außergewöhnlichem Nutzen und Leistung"
spec:
fullDescription: |
**WICHTIGER HINWEIS**
Dies ist eine Shared App. Nach Installation durch den Olares-Admin können alle Benutzer im Cluster sie über die Referenz-App nutzen.
Der Benutzerverkehr geht über den Proxy zu Ollama.

**MODELL**
Verwendet das Ollama-Modell qwen3.5:9b.

Qwen3.5 stellt einen bedeutenden Fortschritt dar und integriert Durchbrüche in multimodalen Lernen, architektonischer Effizienz, Reinforcement-Learning-Skalierung und globaler Zugänglichkeit, um Entwickler und Unternehmen mit beispielloser Fähigkeit und Effizienz zu befähigen.

Qwen3.5 bietet folgende Verbesserungen:

Unified Vision-Language Foundation: Early-Fusion-Training auf multimodalen Tokens erreicht generationsübergreifende Parität mit Qwen3 und übertrifft Qwen3-VL-Modelle bei Reasoning, Coding, Agents und visuellem Verständnis.

Efficient Hybrid Architecture: Gated Delta Networks kombiniert mit Sparse Mixture-of-Experts liefern High-Throughput-Inferenz mit minimaler Latenz und Kosten.

Scalable RL Generalization: Reinforcement Learning skaliert über Millionen-Agenten-Umgebungen mit progressiv komplexen Aufgabenverteilungen für robuste reale Anpassungsfähigkeit.

Global Linguistic Coverage: Erweiterte Unterstützung auf 201 Sprachen und Dialekte für inklusiven weltweiten Einsatz mit nuancenreichem kulturellem und regionalem Verständnis.

Next-Generation Training Infrastructure: Nahezu 100 % multimodale Trainingseffizienz im Vergleich zu Text-only-Training und asynchrone RL-Frameworks für massiv skalierte Agent-Scaffolds und Umgebungsorchestrierung.
upgradeDescription: |
Ollama-Runtime auf ollama/ollama:0.30.10 aktualisiert.
Olares-API-Proxy auf beclab/harveyff-olares-ollama:v0.1.22 aktualisiert (OpenAI-Vision image_url-Weiterleitung).
28 changes: 28 additions & 0 deletions ollamaqwen359bv2/i18n/es-ES/OlaresManifest.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,28 @@
metadata:
title: Qwen3.5 9B Q4_K_M (Ollama)
description: "Modelos multimodales de código abierto que ofrecen utilidad y rendimiento excepcionales"
spec:
fullDescription: |
**NOTA IMPORTANTE**
Esta es una app compartida. Una vez instalada por el administrador de Olares, todos los usuarios del clúster pueden usarla mediante la app de referencia.
El tráfico de usuarios pasa por el proxy hacia Ollama.

**MODELO**
Usa el modelo de Ollama qwen3.5:9b.

Qwen3.5 representa un salto significativo, integrando avances en aprendizaje multimodal, eficiencia arquitectónica, escala de aprendizaje por refuerzo y accesibilidad global para capacitar a desarrolladores y empresas con una capacidad y eficiencia sin precedentes.

Qwen3.5 incluye las siguientes mejoras:

Unified Vision-Language Foundation: el entrenamiento de fusión temprana sobre tokens multimodales alcanza paridad intergeneracional con Qwen3 y supera a los modelos Qwen3-VL en razonamiento, coding, agentes y comprensión visual.

Efficient Hybrid Architecture: Gated Delta Networks combinadas con Mixture-of-Experts dispersas ofrecen inferencia de alto rendimiento con latencia y coste mínimos.

Scalable RL Generalization: aprendizaje por refuerzo escalado en entornos de un millón de agentes con distribuciones de tareas progresivamente complejas para una adaptabilidad robusta en el mundo real.

Global Linguistic Coverage: soporte ampliado a 201 idiomas y dialectos, permitiendo un despliegue mundial inclusivo con comprensión cultural y regional matizada.

Next-Generation Training Infrastructure: eficiencia de entrenamiento multimodal cercana al 100 % frente al entrenamiento solo texto y frameworks RL asíncronos que soportan andamiajes de agentes a gran escala y orquestación de entornos.
upgradeDescription: |
Runtime de Ollama actualizado a ollama/ollama:0.30.10.
Proxy API de Olares actualizado a beclab/harveyff-olares-ollama:v0.1.22 (reenvío de image_url de visión OpenAI).
28 changes: 28 additions & 0 deletions ollamaqwen359bv2/i18n/fr-FR/OlaresManifest.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,28 @@
metadata:
title: Qwen3.5 9B Q4_K_M (Ollama)
description: "Modèles multimodaux open-source offrant une utilité et des performances exceptionnelles"
spec:
fullDescription: |
**NOTE IMPORTANTE**
Il s'agit d'une application partagée. Une fois installée par l'administrateur Olares, tous les utilisateurs du cluster peuvent l'utiliser via l'application de référence.
Le trafic utilisateur passe par le proxy vers Ollama.

**MODÈLE**
Utilise le modèle Ollama qwen3.5:9b.

Qwen3.5 représente une avancée significative, intégrant des percées en apprentissage multimodal, efficacité architecturale, échelle d'apprentissage par renforcement et accessibilité mondiale pour doter développeurs et entreprises d'une capacité et d'une efficacité sans précédent.

Qwen3.5 comporte les améliorations suivantes :

Unified Vision-Language Foundation : l'entraînement early-fusion sur tokens multimodaux atteint la parité intergénérationnelle avec Qwen3 et surpasse les modèles Qwen3-VL en raisonnement, coding, agents et compréhension visuelle.

Efficient Hybrid Architecture : Gated Delta Networks combinées à des Mixture-of-Experts sparses offrent une inférence à haut débit avec une latence et un coût minimaux.

Scalable RL Generalization : apprentissage par renforcement mis à l'échelle sur des environnements à un million d'agents avec des distributions de tâches progressivement complexes pour une adaptabilité robuste au monde réel.

Global Linguistic Coverage : prise en charge élargie à 201 langues et dialectes, permettant un déploiement mondial inclusif avec une compréhension culturelle et régionale nuancée.

Next-Generation Training Infrastructure : efficacité d'entraînement multimodal proche de 100 % par rapport au texte seul et frameworks RL asynchrones supportant des scaffolds d'agents à grande échelle et l'orchestration d'environnements.
upgradeDescription: |
Runtime Ollama mis à jour vers ollama/ollama:0.30.10.
Proxy API Olares mis à jour vers beclab/harveyff-olares-ollama:v0.1.22 (transfert image_url vision OpenAI).
28 changes: 28 additions & 0 deletions ollamaqwen359bv2/i18n/it-IT/OlaresManifest.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,28 @@
metadata:
title: Qwen3.5 9B Q4_K_M (Ollama)
description: "Modelli multimodali open-source che offrono utilità e prestazioni eccezionali"
spec:
fullDescription: |
**NOTA IMPORTANTE**
Questa è un'app condivisa. Una volta installata dall'amministratore Olares, tutti gli utenti del cluster possono usarla tramite l'app di riferimento.
Il traffico utente passa dal proxy verso Ollama.

**MODELLO**
Usa il modello Ollama qwen3.5:9b.

Qwen3.5 rappresenta un significativo passo avanti, integrando progressi nell'apprendimento multimodale, nell'efficienza architetturale, nella scala del reinforcement learning e nell'accessibilità globale per potenziare sviluppatori e imprese con capacità ed efficienza senza precedenti.

Qwen3.5 include i seguenti miglioramenti:

Unified Vision-Language Foundation: il training early-fusion su token multimodali raggiunge la parità intergenerazionale con Qwen3 e supera i modelli Qwen3-VL in reasoning, coding, agent e comprensione visiva.

Efficient Hybrid Architecture: Gated Delta Networks combinate con Mixture-of-Experts sparse offrono inferenza ad alto throughput con latenza e costi minimi.

Scalable RL Generalization: reinforcement learning scalato su ambienti con milioni di agenti e distribuzioni di task progressivamente complesse per una robusta adattabilità reale.

Global Linguistic Coverage: supporto esteso a 201 lingue e dialetti, per un deployment mondiale inclusivo con comprensione culturale e regionale sfumata.

Next-Generation Training Infrastructure: efficienza di training multimodale vicina al 100% rispetto al solo testo e framework RL asincroni che supportano scaffold agent su larga scala e orchestrazione degli ambienti.
upgradeDescription: |
Runtime Ollama aggiornato a ollama/ollama:0.30.10.
Proxy API Olares aggiornato a beclab/harveyff-olares-ollama:v0.1.22 (inoltro image_url vision OpenAI).
28 changes: 28 additions & 0 deletions ollamaqwen359bv2/i18n/ja-JP/OlaresManifest.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,28 @@
metadata:
title: Qwen3.5 9B Q4_K_M (Ollama)
description: "卓越した実用性と性能を提供するオープンソース・マルチモーダルモデル"
spec:
fullDescription: |
**重要なお知らせ**
これは共有アプリです。Olares 管理者がインストールすると、クラスタ内の全ユーザーが参照アプリ経由で利用できます。
ユーザートラフィックはプロキシ経由で Ollama に送られます。

**モデル**
Ollama モデル qwen3.5:9b を使用します。

Qwen3.5 は、マルチモーダル学習、アーキテクチャ効率、強化学習のスケール、グローバルなアクセシビリティにおけるブレークスルーを統合し、開発者と企業に前例のない能力と効率を提供する大きな飛躍です。

Qwen3.5 の強化点:

Unified Vision-Language Foundation:マルチモーダルトークンでの early fusion 学習により Qwen3 と世代横断で同等になり、推論・コーディング・エージェント・視覚理解ベンチマークで Qwen3-VL を上回ります。

Efficient Hybrid Architecture:Gated Delta Networks とスパース Mixture-of-Experts の組み合わせにより、低遅延・低コストで高スループット推論を実現。

Scalable RL Generalization:百万エージェント規模の環境で段階的に複雑なタスク分布に対する強化学習をスケールし、実世界への頑健な適応を実現。

Global Linguistic Coverage:201 の言語・方言に拡張し、文化・地域のニュアンスを踏まえた包括的な世界展開を可能に。

Next-Generation Training Infrastructure:テキストのみ学習に対してほぼ 100% のマルチモーダル学習効率と、大規模エージェント足場および環境オーケストレーションを支える非同期 RL フレームワーク。
upgradeDescription: |
Ollama ランタイムを ollama/ollama:0.30.10 にアップグレード。
Olares API プロキシを beclab/harveyff-olares-ollama:v0.1.22 にアップグレード(OpenAI vision image_url 転送)。
2 changes: 1 addition & 1 deletion ollamaqwen359bv2/ollamaqwen359bv2/Chart.yaml
Original file line number Diff line number Diff line change
@@ -1,5 +1,5 @@
apiVersion: v2
appVersion: '1.25.3-2'
appVersion: '1.29.2.5'
description: description
name: ollamaqwen359bv2
type: application
Expand Down
32 changes: 21 additions & 11 deletions ollamaqwen359bv2/ollamaqwen359bv2/templates/clientproxy.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -3,10 +3,14 @@ apiVersion: v1
data:
nginx.conf: |
server {

client_body_temp_path /tmp/client_body;
proxy_temp_path /tmp/proxy;
fastcgi_temp_path /tmp/fastcgi;
uwsgi_temp_path /tmp/uwsgi;
scgi_temp_path /tmp/scgi;
listen 8080;
access_log /opt/bitnami/openresty/nginx/logs/access.log;
error_log /opt/bitnami/openresty/nginx/logs/error.log;
access_log /dev/stdout;
error_log /dev/stderr;

proxy_connect_timeout 600s;
proxy_send_timeout 600s;
Expand Down Expand Up @@ -60,15 +64,22 @@ spec:
items:
- key: nginx.conf
path: nginx.conf
- name: nginx-runtime
emptyDir: {}
containers:
- name: nginx
image: "docker.io/beclab/aboveos-bitnami-openresty:1.25.3-2"
image: "docker.io/openresty/openresty:1.29.2.5-bookworm-fat"
command:
- /usr/local/openresty/bin/openresty
args:
- -g
- "daemon off; pid /tmp/nginx.pid;"
securityContext:
runAsUser: 1000
runAsGroup: 1000
ports:
- containerPort: 8080
protocol: TCP
env:
- name: OPENRESTY_CONF_FILE
value: /etc/nginx/nginx.conf
startupProbe:
tcpSocket:
port: 8080
Expand All @@ -92,11 +103,10 @@ spec:
memory: 64Mi
volumeMounts:
- name: nginx-config
mountPath: /etc/nginx/nginx.conf
subPath: nginx.conf
- name: nginx-config
mountPath: /opt/bitnami/openresty/nginx/conf/server_blocks/nginx.conf
mountPath: /etc/nginx/conf.d/default.conf
subPath: nginx.conf
- name: nginx-runtime
mountPath: /var/run/openresty

---
apiVersion: v1
Expand Down
2 changes: 1 addition & 1 deletion ollamaqwen359bv2/owners
Original file line number Diff line number Diff line change
Expand Up @@ -3,4 +3,4 @@ owners:
- 'TShentu'
- 'hysyeah'
- 'pengpeng'
- 'harveyff'
- 'harveyff'