apunkt/plano
1
0
Fork 0
mirror of https://github.com/katanemo/plano.git synced 2026-05-05 22:02:43 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions

Adding support for wildcard models in the model_providers config (#696)

Browse source 
* cleaning up plano cli commands

* adding support for wildcard model providers

* fixing compile errors

* fixing bugs related to default model provider, provider hint and duplicates in the model provider list

* fixed cargo fmt issues

* updating tests to always include the model id

* using default for the prompt_gateway path

* fixed the model name, as gpt-5-mini-2025-08-07 wasn't in the config

* making sure that all aliases and models match the config

* fixed the config generator to allow for base_url providers LLMs to include wildcard models

* re-ran the models list utility and added a shell script to run it

* updating docs to mention wildcard model providers

* updated provider_models.json to yaml, added that file to our docs for reference

* updating the build docs to use the new root-based build

---------

Co-authored-by: Salman Paracha <salmanparacha@MacBook-Pro-342.local>
This commit is contained in:
Salman Paracha 2026-01-28 17:47:33 -08:00 • committed by GitHub
parent 8428b06e22
commit 2941392ed1
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
42 changed files with 1748 additions and 202 deletions
Download patch file Download diff file Expand all files Collapse all files

64
cli/planoai/config_generator.py
View file

@ -187,11 +187,21 @@ def validate_and_render_schema():
model_name = model_provider.get("model")
print("Processing model_provider: ", model_provider)
if model_name in model_name_keys:
# Check if this is a wildcard model (provider/*)
is_wildcard = False
if "/" in model_name:
model_name_tokens = model_name.split("/")
if len(model_name_tokens) >= 2 and model_name_tokens[-1] == "*":
is_wildcard = True
if model_name in model_name_keys and not is_wildcard:
raise Exception(
f"Duplicate model name {model_name}, please provide unique model name for each model_provider"
)
model_name_keys.add(model_name)
if not is_wildcard:
model_name_keys.add(model_name)
if model_provider.get("name") is None:
model_provider["name"] = model_name
@ -200,9 +210,23 @@ def validate_and_render_schema():
model_name_tokens = model_name.split("/")
if len(model_name_tokens) < 2:
raise Exception(
f"Invalid model name {model_name}. Please provide model name in the format <provider>/<model_id>."
f"Invalid model name {model_name}. Please provide model name in the format <provider>/<model_id> or <provider>/* for wildcards."
)
provider = model_name_tokens[0]
provider = model_name_tokens[0].strip()
# Check if this is a wildcard (provider/*)
is_wildcard = model_name_tokens[-1].strip() == "*"
# Validate wildcard constraints
if is_wildcard:
if model_provider.get("default", False):
raise Exception(
f"Model {model_name} is configured as default but uses wildcard (*). Default models cannot be wildcards."
)
if model_provider.get("routing_preferences"):
raise Exception(
f"Model {model_name} has routing_preferences but uses wildcard (*). Models with routing preferences cannot be wildcards."
)
# Validate azure_openai and ollama provider requires base_url
if (provider in SUPPORTED_PROVIDERS_WITH_BASE_URL) and model_provider.get(
@ -213,7 +237,9 @@ def validate_and_render_schema():
)
model_id = "/".join(model_name_tokens[1:])
if provider not in SUPPORTED_PROVIDERS:
# For wildcard providers, allow any provider name
if not is_wildcard and provider not in SUPPORTED_PROVIDERS:
if (
model_provider.get("base_url", None) is None
or model_provider.get("provider_interface", None) is None
@ -222,16 +248,32 @@ def validate_and_render_schema():
f"Must provide base_url and provider_interface for unsupported provider {provider} for model {model_name}. Supported providers are: {', '.join(SUPPORTED_PROVIDERS)}"
)
provider = model_provider.get("provider_interface", None)
elif model_provider.get("provider_interface", None) is not None:
elif is_wildcard and provider not in SUPPORTED_PROVIDERS:
# Wildcard models with unsupported providers require base_url and provider_interface
if (
model_provider.get("base_url", None) is None
or model_provider.get("provider_interface", None) is None
):
raise Exception(
f"Must provide base_url and provider_interface for unsupported provider {provider} for wildcard model {model_name}. Supported providers are: {', '.join(SUPPORTED_PROVIDERS)}"
)
provider = model_provider.get("provider_interface", None)
elif (
provider in SUPPORTED_PROVIDERS
and model_provider.get("provider_interface", None) is not None
):
# For supported providers, provider_interface should not be manually set
raise Exception(
f"Please provide provider interface as part of model name {model_name} using the format <provider>/<model_id>. For example, use 'openai/gpt-3.5-turbo' instead of 'gpt-3.5-turbo' "
)
if model_id in model_name_keys:
raise Exception(
f"Duplicate model_id {model_id}, please provide unique model_id for each model_provider"
)
model_name_keys.add(model_id)
# For wildcard models, don't add model_id to the keys since it's "*"
if not is_wildcard:
if model_id in model_name_keys:
raise Exception(
f"Duplicate model_id {model_id}, please provide unique model_id for each model_provider"
)
model_name_keys.add(model_id)
for routing_preference in model_provider.get("routing_preferences", []):
if routing_preference.get("name") in model_usage_name_keys:

2
cli/planoai/core.py
View file

@ -145,7 +145,7 @@ def stop_docker_container(service=PLANO_DOCKER_NAME):
def start_cli_agent(arch_config_file=None, settings_json="{}"):
"""Start a CLI client connected to Arch."""
"""Start a CLI client connected to Plano."""
with open(arch_config_file, "r") as file:
arch_config = file.read()

12
cli/planoai/main.py
View file

@ -74,7 +74,7 @@ def main(ctx, version):
log.info(f"Starting plano cli version: {get_version()}")
if ctx.invoked_subcommand is None:
click.echo("""Arch (The Intelligent Prompt Gateway) CLI""")
click.echo("""Plano (AI-native proxy and dataplane for agentic apps) CLI""")
click.echo(logo)
click.echo(ctx.get_help())
@ -121,16 +121,16 @@ def build():
@click.command()
@click.argument("file", required=False) # Optional file argument
@click.option(
"--path", default=".", help="Path to the directory containing arch_config.yaml"
"--path", default=".", help="Path to the directory containing config.yaml"
)
@click.option(
"--foreground",
default=False,
help="Run Arch in the foreground. Default is False",
help="Run Plano in the foreground. Default is False",
is_flag=True,
)
def up(file, path, foreground):
"""Starts Arch."""
"""Starts Plano."""
# Use the utility function to find config file
arch_config_file = find_config_file(path, file)
@ -270,7 +270,7 @@ def logs(debug, follow):
help="Additional settings as JSON string for the CLI agent.",
)
def cli_agent(type, file, path, settings):
"""Start a CLI agent connected to Arch.
"""Start a CLI agent connected to Plano.
CLI_AGENT: The type of CLI agent to start (currently only 'claude' is supported)
"""
@ -278,7 +278,7 @@ def cli_agent(type, file, path, settings):
# Check if plano docker container is running
archgw_status = docker_container_status(PLANO_DOCKER_NAME)
if archgw_status != "running":
log.error(f"archgw docker container is not running (status: {archgw_status})")
log.error(f"plano docker container is not running (status: {archgw_status})")
log.error("Please start plano using the 'planoai up' command.")
sys.exit(1)