kalavai-client 0.5.20__py3-none-any.whl → 0.5.22__py3-none-any.whl

kalavai_client/__init__.py

@@ -1,2 +1,2 @@
 
-__version__ = "0.5.20"
+__version__ = "0.5.22"

kalavai_client/assets/apps.yaml

@@ -139,7 +139,7 @@ releases:
     - name: replicas
       value: 2
     - name: image_tag
-      value: "v2025.01.10"
+      value: "v2025.03.1"
     - name: deployment.in_cluster
       value: "True"
     - name: deployment.use_auth_key

kalavai_client/assets/docker-compose-template.yaml

@@ -103,7 +103,7 @@ services:
   {% endif %}
 {% endif %}
 
-networks:
-  custom-network:
-    driver: bridge
+# networks:
+#   custom-network:
+#     driver: bridge
 

kalavai_client/bridge_api.py

@@ -64,7 +64,8 @@ def pool_join(request: JoinPoolRequest):
     result = join_pool(
         token=request.token,
         num_gpus=request.num_gpus,
-        node_name=request.node_name
+        node_name=request.node_name,
+        ip_address=request.ip_address
     )
     return result
 

kalavai_client/bridge_models.py

@@ -17,6 +17,7 @@ class DeleteNodesRequest(BaseModel):
 
 class JoinPoolRequest(BaseModel):
     token: str
+    ip_address: str = None
     node_name: str = None
     num_gpus: int = None
 

kalavai_client/cli.py

@@ -319,14 +319,12 @@ def pool__publish(*others, description=None):
         console.log(f"[red]Problems with your pool: {str(e)}")
         return
     choices = select_token_type()
-    token = pool__token(**choices)
+    token = pool__token(**choices)["token"]
     
     if description is None:
         console.log("[yellow] [Markdown] In a few words (max 500 chars), describe your goals with this cluster. Remember, this is what other users will see to decide whether to share their resources with you, [blue]so inspire them!")
         description = input(f"(You can edit this later in {KALAVAI_PLATFORM_URL}\n")
     
-    description = description
-    
     try:
         valid = check_token(token=token, public=True)
         if "error" in valid:
@@ -462,7 +460,7 @@ def pool__token(*others, admin=False, user=False, worker=False):
         console.log(f"[red]{join_token}")
     else:
         console.log("[green]Join token:")
-        print(join_token)
+        print(join_token["token"])
     return join_token
 
 @arguably.command
@@ -513,11 +511,16 @@ def pool__join(token, *others, node_name=None):
         console.log("[red]Installation aborted")
         return
     
+    # select IP address (for external discovery)
+    console.log(f"Scanning for valid IPs")
+    ip_address = select_ip_address()
+    
     console.log("Connecting worker to the pool...")
     result = join_pool(
         token=token,
         node_name=node_name,
-        num_gpus=num_gpus
+        num_gpus=num_gpus,
+        ip_address=ip_address
     )
     if "error" in result:
         console.log(f"[red]Error when connecting: {result}")
@@ -1048,7 +1051,7 @@ def job__run(template_name, *others, values: str=None, force_namespace: str=None
     )
 
     if "error" in result:
-        console.log(f"[red]Error when deploying job: {str(e)}")
+        console.log(f"[red]Error when deploying job: {str(result['error'])}")
     else:
         console.log(f"[green]{template_name} job deployed")
 

kalavai_client/core.py

@@ -267,6 +267,8 @@ def fetch_job_details(jobs: list[Job]):
                 if ns != namespace: # same job name, different namespace
                     continue
                 for _, values in ss.items():
+                    # TODO get 'restart_count' from values['conditions'][-1]["restart_count"]
+                    # TODO: get nodes involved in deployment (needs kubewatcher)
                     workers_status[values["status"]] += 1
             workers = "\n".join([f"{k}: {v}" for k, v in workers_status.items()])
             # get URL details
@@ -287,10 +289,14 @@ def fetch_job_details(jobs: list[Job]):
             urls = [f"http://{load_server_info(data_key=SERVER_IP_KEY, file=USER_LOCAL_SERVER_FILE)}:{node_port}" for node_port in node_ports]
             if "Ready" in workers_status and len(workers_status) == 1:
                 status = "running"
-            elif any([st in workers_status for st in ["Failed", "Completed"]]):
+            elif any([st in workers_status for st in ["Failed"]]):
                 status = "error"
-            else:
+            elif any([st in workers_status for st in ["Pending"]]) or len(workers_status) == 0:
                 status = "pending"
+            elif any([st in workers_status for st in ["Succeeded", "Completed"]]):
+                status = "completed"
+            else:
+                status = "working"
             job_details.append(
                 Job(owner=namespace,
                     name=deployment,
@@ -567,7 +573,7 @@ def get_max_gpus():
     except:
         return 0
 
-def join_pool(token, num_gpus=None, node_name=None):
+def join_pool(token, num_gpus=None, node_name=None, ip_address=None):
     compatibility = check_worker_compatibility()
     if len(compatibility["issues"]) > 0:
         return {"error": compatibility["issues"]}
@@ -624,6 +630,7 @@ def join_pool(token, num_gpus=None, node_name=None):
     # Generate docker compose recipe
     generate_compose_config(
         role="agent",
+        node_ip_address=ip_address,
         pool_ip=f"https://{kalavai_seed_ip}:6443",
         pool_token=kalavai_token,
         num_gpus=num_gpus,
@@ -699,6 +706,9 @@ def create_pool(cluster_name: str, ip_address: str, app_values: str=None, pool_c
             node_labels[USER_NODE_LABEL] = user["username"]
         except Exception as e:
             return {"error": f"[red]Error when joining network: {str(e)}"}
+        
+    if num_gpus is None:
+        num_gpus = get_max_gpus()
     
     # Generate docker compose recipe
     generate_compose_config(

{kalavai_client-0.5.20.dist-info → kalavai_client-0.5.22.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: kalavai-client
-Version: 0.5.20
+Version: 0.5.22
 Summary: Client app for kalavai platform
 License: Apache-2.0
 Keywords: LLM,platform
@@ -26,6 +26,7 @@ Requires-Dist: Pillow (==10.3.0)
 Requires-Dist: anvil-uplink (==0.5.1)
 Requires-Dist: arguably (>=1.2.5)
 Requires-Dist: build ; extra == "dev"
+Requires-Dist: fastapi (==0.115.8)
 Requires-Dist: importlib_resources (==6.5.2)
 Requires-Dist: jinja2 (==3.1.4)
 Requires-Dist: netifaces (==0.11.0)
@@ -38,6 +39,7 @@ Requires-Dist: requests (>=2.25)
 Requires-Dist: rich (==13.7.1)
 Requires-Dist: setuptools (>75.0.0)
 Requires-Dist: twine ; extra == "dev"
+Requires-Dist: uvicorn (==0.34.0)
 Project-URL: Homepage, https://platform.kalavai.net
 Project-URL: Website, https://kalavai.net
 Description-Content-Type: text/markdown
@@ -154,8 +156,6 @@ Not what you were looking for? [Tell us](https://github.com/kalavai-net/kalavai-
 
 The `kalavai-client` is the main tool to interact with the Kalavai platform, to create and manage both local and public pools and also to interact with them (e.g. deploy models). Let's go over its installation. 
 
-From release **v0.5.0, you can now install `kalavai-client` in non-worker computers**. You can run a pool on a set of machines and have the client on a remote computer from which you access the LLM pool. Because the client only requires having python installed, this means more computers are now supported to run it.
-
 
 <details>
 
@@ -211,7 +211,7 @@ pip install kalavai-client
 ```
 
 
-## Createa a local, private LLM pool
+## Create a a local, private LLM pool
 
 > Kalavai is **free to use, no caps, for both commercial and non-commercial purposes**. All you need to get started is one or more computers that can see each other (i.e. within the same network), and you are good to go. If you are interested in join computers in different locations / networks, [contact us](mailto:info@kalavai.net) or [book a demo](https://app.onecal.io/b/kalavai/book-a-demo) with the founders.
 
@@ -237,11 +237,7 @@ Check out [our guide](https://kalavai-net.github.io/kalavai-client/public_llm_po
 
 ## Enough already, let's run stuff!
 
-Check our [examples](examples/) to put your new AI pool to good use!
-- [Single node vLLM GPU LLM](examples/singlenode_gpu_vllm.md) deployment
-- [Multi node vLLM GPU LLM](examples/multinode_gpu_vllm.md) deployment
-- [Aphrodite-engine quantized LLM](examples/quantized_gpu_llm.md) deployment, including Kobold interface
-- [Ray cluster](examples/ray_cluster.md) for distributed computation.
+Check our [examples](examples/) to put your new AI pool to good use! For an end to end tour, check our [self-hosted](https://kalavai-net.github.io/kalavai-client/self_hosted_llm_pool/) and [public LLM pools](https://kalavai-net.github.io/kalavai-client/public_llm_pool/) guides.
 
 
 ## Compatibility matrix

{kalavai_client-0.5.20.dist-info → kalavai_client-0.5.22.dist-info}/RECORD

@@ -1,25 +1,25 @@
-kalavai_client/__init__.py,sha256=YJ16sMq9PK__3-NzHbGxz6Zz5IYwpCEzy4p-Yddy-o8,23
+kalavai_client/__init__.py,sha256=B8BZXxq8ga04_YWJxV1IYhwDos_rr2WoHNu4_srpzOM,23
 kalavai_client/__main__.py,sha256=WQUfxvRsBJH5gsCJg8pLz95QnZIj7Ol8psTO77m0QE0,73
 kalavai_client/assets/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-kalavai_client/assets/apps.yaml,sha256=V1x1FY-fyYsYrXvcIMv3QrBCgJ7jNunluRyJh67eWB0,5983
+kalavai_client/assets/apps.yaml,sha256=yeKtxkz1e_EW7tNpxMb0M-FAF4vKLW27nMRd8xJY9C8,5982
 kalavai_client/assets/apps_values.yaml,sha256=CjKVelPQHd-hm-DTMEuya92feKiphU9mh3HrosLYYPE,1676
 kalavai_client/assets/docker-compose-gui.yaml,sha256=oE-neG3HC2PhdH-mIxrZdQlhqnycrSP_p8fRxjaxDFE,192
-kalavai_client/assets/docker-compose-template.yaml,sha256=Fnsk44oFtUqS2E77umVUiWl1-eS96E13TPe31GYo1XI,2825
+kalavai_client/assets/docker-compose-template.yaml,sha256=ii24Nn-dM5cZk9lxFgrzxnmK7yv_6kIIw7KUlWhvYeI,2831
 kalavai_client/assets/nginx.conf,sha256=drVVCg8GHucz7hmt_BI6giAhK92OV71257NTs3LthwM,225
 kalavai_client/assets/pool_config_template.yaml,sha256=fFz4w2-fMKD5KvyzFdfcWD_jSneRlmnjLc8hCctweX0,576
 kalavai_client/assets/pool_config_values.yaml,sha256=VrM3XHQfQo6QLZ68qvagooUptaYgl1pszniY_JUtemk,233
 kalavai_client/assets/user_workspace.yaml,sha256=wDvlMYknOPABAEo0dsQwU7bac8iubjAG9tdkFbJZ5Go,476
 kalavai_client/assets/user_workspace_values.yaml,sha256=G0HOzQUxrDMCwuW9kbWUZaKMzDDPVwDwzBHCL2Xi2ZM,542
 kalavai_client/auth.py,sha256=QsBh28L2LwjBBK6pTUE4Xu36lLDTyetyU1YfS1Hbb6g,1717
-kalavai_client/bridge_api.py,sha256=hp5YjMu0HBI9VGMx6hahXfMIGPLwNtSd09UKxmKnGXc,4852
-kalavai_client/bridge_models.py,sha256=rXBnE5r6Oe9GxGkk1ITkvp6YQqahp72Rrzf-QM2quH8,771
-kalavai_client/cli.py,sha256=u9zy2H3Ntn3fOnQyYU8XjVPmouZNqyJxiyXBVA5EtEA,47862
+kalavai_client/bridge_api.py,sha256=lFS1fvH7UQxZ24vrHGeea5z8bnGWhOZhgU_zKIrnlHI,4891
+kalavai_client/bridge_models.py,sha256=hWxEODqHwc8Sj21zkIZoZmU9W3PvNdK-CA1tH6yMQ4s,798
+kalavai_client/cli.py,sha256=8HBAVAgLEJo6XHHF5eORIvtn7_ObykkCQZwbP5ngzpU,48024
 kalavai_client/cluster.py,sha256=gwjmdsd--YrffT0BmZDOEpbrdm3lPskUuN5jdgcrOR0,12947
-kalavai_client/core.py,sha256=JTyGpi4Xez_kAgEpcInErS-Ot3dFLoNY48nEBQJLuOc,29653
+kalavai_client/core.py,sha256=c9UI_RJ7QmvWPfB9VQpc5QtNCbExOcPtGc-zed1ZgzI,30187
 kalavai_client/env.py,sha256=Zg2pP-xGJpQumo56KMBxBLgIsBmcNN0S9R-ZP2-s630,2604
 kalavai_client/utils.py,sha256=rz5W9PRZrTpgdmOs6yeqUi4f_q_L-3BJ5g1o7Asgnyo,13386
-kalavai_client-0.5.20.dist-info/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
-kalavai_client-0.5.20.dist-info/METADATA,sha256=aHwXRZpoiBBSDbjthDxR2ncdeLTInAUJ2cci3Li7Dn0,14800
-kalavai_client-0.5.20.dist-info/WHEEL,sha256=XbeZDeTWKc1w7CSIyre5aMDU_-PohRwTQceYnisIYYY,88
-kalavai_client-0.5.20.dist-info/entry_points.txt,sha256=9T6D45gxwzfVbglMm1r6XPdXuuZdHfy_7fCeu2jUphc,50
-kalavai_client-0.5.20.dist-info/RECORD,,
+kalavai_client-0.5.22.dist-info/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
+kalavai_client-0.5.22.dist-info/METADATA,sha256=fSOCPN-YiifsInCwIyooSmPZ6GsLdqmANN6s7rtHKNg,14443
+kalavai_client-0.5.22.dist-info/WHEEL,sha256=XbeZDeTWKc1w7CSIyre5aMDU_-PohRwTQceYnisIYYY,88
+kalavai_client-0.5.22.dist-info/entry_points.txt,sha256=9T6D45gxwzfVbglMm1r6XPdXuuZdHfy_7fCeu2jUphc,50
+kalavai_client-0.5.22.dist-info/RECORD,,