ollama-helm
Ollama Helm Chart
Ollama, get up and running with large language models, locally.
This Chart is for deploying Ollama.
Requirements
-
Kubernetes:
>= 1.16.0-0for CPU only -
Kubernetes:
>= 1.26.0-0for GPU stable support (NVIDIA and AMD)
Not all GPUs are currently supported with ollama (especially with AMD)
Deploying Ollama chart
To install the ollama chart in the ollama namespace:
helm repo add ollama-helm https://otwld.github.io/ollama-helm/
helm repo update
helm install ollama ollama-helm/ollama --namespace ollama
Upgrading Ollama chart
First please read the release notes of Ollama to make sure there are no backwards incompatible changes.
Make adjustments to your values as needed, then run helm upgrade:
# -- This pulls the latest version of the ollama chart from the repo.
helm repo update
helm upgrade ollama ollama-helm/ollama --namespace ollama --values values.yaml
Uninstalling Ollama chart
To uninstall/delete the ollama deployment in the ollama namespace:
helm delete ollama --namespace ollama
Substitute your values if they differ from the examples. See helm delete --help for a full reference on delete parameters and flags.
Interact with Ollama
- Ollama documentation can be found HERE
- Interact with RESTful API: Ollama API
- Interact with official clients libraries: ollama-js and ollama-python
- Interact with langchain: langchain-js and langchain-python
Examples
- It’s highly recommended to run an updated version of Kubernetes for deploying ollama with GPU
Basic values.yaml example with GPU and two models pull at startup
ollama:
gpu:
# -- Enable GPU integration
enabled: true
# -- GPU type: 'nvidia' or 'amd'
type: 'nvidia'
# -- Specify the number of GPU to 2
number: 2
# -- List of models to pull at container startup
models:
- mistral
- llama2
Basic values.yaml example with Ingress
ollama:
models:
- llama2
ingress:
enabled: true
hosts:
- host: ollama.domain.lan
paths:
- path: /
pathType: Prefix
- API is now reachable at
ollama.domain.lan
Helm Values
- See values.yaml to see the Chart’s default values.
| Key | Type | Default | Description |
|—–|——|———|————-|
| affinity | object | {} | Affinity for pod assignment |
| autoscaling.enabled | bool | false | Enable autoscaling |
| autoscaling.maxReplicas | int | 100 | Number of maximum replicas |
| autoscaling.minReplicas | int | 1 | Number of minimum replicas |
| autoscaling.targetCPUUtilizationPercentage | int | 80 | CPU usage to target replica |
| extraArgs | list | [] | Additional arguments on the output Deployment definition. |
| extraEnv | list | [] | Additional environments variables on the output Deployment definition. |
| fullnameOverride | string | "" | String to fully override template |
| image.pullPolicy | string | "IfNotPresent" | Docker pull policy |
| image.repository | string | "ollama/ollama" | Docker image registry |
| image.tag | string | "" | Docker image tag, overrides the image tag whose default is the chart appVersion. |
| imagePullSecrets | list | [] | Docker registry secret names as an array |
| ingress.annotations | object | {} | Additional annotations for the Ingress resource. |
| ingress.className | string | "" | IngressClass that will be used to implement the Ingress (Kubernetes 1.18+) |
| ingress.enabled | bool | false | Enable ingress controller resource |
| ingress.hosts[0].host | string | "ollama.local" | |
| ingress.hosts[0].paths[0].path | string | "/" | |
| ingress.hosts[0].paths[0].pathType | string | "Prefix" | |
| ingress.tls | list | [] | The tls configuration for hostnames to be covered with this ingress record. |
| livenessProbe.enabled | bool | true | Enable livenessProbe |
| livenessProbe.failureThreshold | int | 6 | Failure threshold for livenessProbe |
| livenessProbe.initialDelaySeconds | int | 60 | Initial delay seconds for livenessProbe |
| livenessProbe.path | string | "/" | Request path for livenessProbe |
| livenessProbe.periodSeconds | int | 10 | Period seconds for livenessProbe |
| livenessProbe.successThreshold | int | 1 | Success threshold for livenessProbe |
| livenessProbe.timeoutSeconds | int | 5 | Timeout seconds for livenessProbe |
| nameOverride | string | "" | String to partially override template (will maintain the release name) |
| nodeSelector | object | {} | Node labels for pod assignment. |
| ollama.gpu.enabled | bool | false | Enable GPU integration |
| ollama.gpu.number | int | 1 | Specify the number of GPU |
| ollama.gpu.type | string | "nvidia" | GPU type: ‘nvidia’ or ‘amd’ If ‘ollama.gpu.enabled’, default value is nvidia If set to ‘amd’, this will add ‘rocm’ suffix to image tag if ‘image.tag’ is not override This is due cause AMD and CPU/CUDA are different images |
| ollama.insecure | bool | false | Add insecure flag for pulling at container startup |
| ollama.models | object | {} | List of models to pull at container startup The more you add, the longer the container will take to start if models are not present models: - llama2 - mistral |
| persistentVolume.accessModes | list | ["ReadWriteOnce"] | Ollama server data Persistent Volume access modes Must match those of existing PV or dynamic provisioner Ref: http://kubernetes.io/docs/user-guide/persistent-volumes/ |
| persistentVolume.annotations | object | {} | Ollama server data Persistent Volume annotations |
| persistentVolume.enabled | bool | true | Enable persistence using PVC |
| persistentVolume.existingClaim | string | "" | If you’d like to bring your own PVC for persisting Ollama state, pass the name of the created + ready PVC here. If set, this Chart will not create the default PVC. Requires server.persistentVolume.enabled: true |
| persistentVolume.size | string | "30Gi" | Ollama server data Persistent Volume size |
| persistentVolume.storageClass | string | "" | Ollama server data Persistent Volume Storage Class If defined, storageClassName:
Support
- For questions, suggestions, and discussion about Ollama please refer to the Ollama issue page
- For questions, suggestions, and discussion about this chart please visite Ollama-Helm issue page