phoenix-oss/llama-stack
1
0
Fork 0
forked from phoenix-oss/llama-stack-mirror
Code Pull requests Releases Packages 2 Activity Actions

Use HF names for registering fireworks and together models

Browse source
This commit is contained in:
Ashwin Bharambe 2024-11-18 22:34:26 -08:00
parent 6765fd76ff
commit 7693786322
7 changed files with 76 additions and 54 deletions
Download patch file Download diff file Expand all files Collapse all files

18
docs/source/getting_started/distributions/self_hosted_distro/fireworks.md
View file

@ -22,15 +22,15 @@ The following environment variables can be configured:
The following models are available by default: The following models are available by default:
- `fireworks/llama-v3p1-8b-instruct` - `meta-llama/Llama-3.1-8B-Instruct (fireworks/llama-v3p1-8b-instruct)`
- `fireworks/llama-v3p1-70b-instruct` - `meta-llama/Llama-3.1-70B-Instruct (fireworks/llama-v3p1-70b-instruct)`
- `fireworks/llama-v3p1-405b-instruct` - `meta-llama/Llama-3.1-405B-Instruct-FP8 (fireworks/llama-v3p1-405b-instruct)`
- `fireworks/llama-v3p2-1b-instruct` - `meta-llama/Llama-3.2-3B-Instruct (fireworks/llama-v3p2-1b-instruct)`
- `fireworks/llama-v3p2-3b-instruct` - `meta-llama/Llama-3.2-11B-Vision-Instruct (fireworks/llama-v3p2-3b-instruct)`
- `fireworks/llama-v3p2-11b-vision-instruct` - `meta-llama/Llama-3.2-11B-Vision-Instruct (fireworks/llama-v3p2-11b-vision-instruct)`
- `fireworks/llama-v3p2-90b-vision-instruct` - `meta-llama/Llama-3.2-90B-Vision-Instruct (fireworks/llama-v3p2-90b-vision-instruct)`
- `fireworks/llama-guard-3-8b` - `meta-llama/Llama-Guard-3-8B (fireworks/llama-guard-3-8b)`
- `fireworks/llama-guard-3-11b-vision` - `meta-llama/Llama-Guard-3-11B-Vision (fireworks/llama-guard-3-11b-vision)`
### Prerequisite: API Keys ### Prerequisite: API Keys

16
docs/source/getting_started/distributions/self_hosted_distro/together.md
View file

@ -22,14 +22,14 @@ The following environment variables can be configured:
The following models are available by default: The following models are available by default:
- `meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo` - `meta-llama/Llama-3.1-8B-Instruct`
- `meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo` - `meta-llama/Llama-3.1-70B-Instruct`
- `meta-llama/Meta-Llama-3.1-405B-Instruct-Turbo` - `meta-llama/Llama-3.1-405B-Instruct-FP8`
- `meta-llama/Llama-3.2-3B-Instruct-Turbo` - `meta-llama/Llama-3.2-3B-Instruct`
- `meta-llama/Llama-3.2-11B-Vision-Instruct-Turbo` - `meta-llama/Llama-3.2-11B-Vision-Instruct`
- `meta-llama/Llama-3.2-90B-Vision-Instruct-Turbo` - `meta-llama/Llama-3.2-90B-Vision-Instruct`
- `meta-llama/Meta-Llama-Guard-3-8B` - `meta-llama/Llama-Guard-3-8B`
- `meta-llama/Llama-Guard-3-11B-Vision-Turbo` - `meta-llama/Llama-Guard-3-11B-Vision`
### Prerequisite: API Keys ### Prerequisite: API Keys

2
llama_stack/templates/fireworks/doc_template.md
View file

@ -20,7 +20,7 @@ The following environment variables can be configured:
The following models are available by default: The following models are available by default:
{% for model in default_models %} {% for model in default_models %}
- `{{ model.model_id }}` - `{{ model.model_id }} ({{ model.provider_model_id }})`
{% endfor %} {% endfor %}
{% endif %} {% endif %}

13
llama_stack/templates/fireworks/fireworks.py
View file

@ -6,6 +6,8 @@
from pathlib import Path from pathlib import Path
from llama_models.sku_list import all_registered_models
from llama_stack.distribution.datatypes import ModelInput, Provider, ShieldInput from llama_stack.distribution.datatypes import ModelInput, Provider, ShieldInput
from llama_stack.providers.remote.inference.fireworks import FireworksImplConfig from llama_stack.providers.remote.inference.fireworks import FireworksImplConfig
from llama_stack.providers.remote.inference.fireworks.fireworks import MODEL_ALIASES from llama_stack.providers.remote.inference.fireworks.fireworks import MODEL_ALIASES
@ -28,7 +30,16 @@ def get_distribution_template() -> DistributionTemplate:
config=FireworksImplConfig.sample_run_config(), config=FireworksImplConfig.sample_run_config(),
) )
default_models = [ModelInput(model_id=m.provider_model_id) for m in MODEL_ALIASES] core_model_to_hf_repo = {
m.descriptor(): m.huggingface_repo for m in all_registered_models()
}
default_models = [
ModelInput(
model_id=core_model_to_hf_repo[m.llama_model],
provider_model_id=m.provider_model_id,
)
for m in MODEL_ALIASES
]
return DistributionTemplate( return DistributionTemplate(
name="fireworks", name="fireworks",

36
llama_stack/templates/fireworks/run.yaml
View file

@ -45,41 +45,41 @@ metadata_store:
db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/fireworks}/registry.db db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/fireworks}/registry.db
models: models:
- metadata: {} - metadata: {}
model_id: fireworks/llama-v3p1-8b-instruct model_id: meta-llama/Llama-3.1-8B-Instruct
provider_id: null provider_id: null
provider_model_id: null provider_model_id: fireworks/llama-v3p1-8b-instruct
- metadata: {} - metadata: {}
model_id: fireworks/llama-v3p1-70b-instruct model_id: meta-llama/Llama-3.1-70B-Instruct
provider_id: null provider_id: null
provider_model_id: null provider_model_id: fireworks/llama-v3p1-70b-instruct
- metadata: {} - metadata: {}
model_id: fireworks/llama-v3p1-405b-instruct model_id: meta-llama/Llama-3.1-405B-Instruct-FP8
provider_id: null provider_id: null
provider_model_id: null provider_model_id: fireworks/llama-v3p1-405b-instruct
- metadata: {} - metadata: {}
model_id: fireworks/llama-v3p2-1b-instruct model_id: meta-llama/Llama-3.2-3B-Instruct
provider_id: null provider_id: null
provider_model_id: null provider_model_id: fireworks/llama-v3p2-1b-instruct
- metadata: {} - metadata: {}
model_id: fireworks/llama-v3p2-3b-instruct model_id: meta-llama/Llama-3.2-11B-Vision-Instruct
provider_id: null provider_id: null
provider_model_id: null provider_model_id: fireworks/llama-v3p2-3b-instruct
- metadata: {} - metadata: {}
model_id: fireworks/llama-v3p2-11b-vision-instruct model_id: meta-llama/Llama-3.2-11B-Vision-Instruct
provider_id: null provider_id: null
provider_model_id: null provider_model_id: fireworks/llama-v3p2-11b-vision-instruct
- metadata: {} - metadata: {}
model_id: fireworks/llama-v3p2-90b-vision-instruct model_id: meta-llama/Llama-3.2-90B-Vision-Instruct
provider_id: null provider_id: null
provider_model_id: null provider_model_id: fireworks/llama-v3p2-90b-vision-instruct
- metadata: {} - metadata: {}
model_id: fireworks/llama-guard-3-8b model_id: meta-llama/Llama-Guard-3-8B
provider_id: null provider_id: null
provider_model_id: null provider_model_id: fireworks/llama-guard-3-8b
- metadata: {} - metadata: {}
model_id: fireworks/llama-guard-3-11b-vision model_id: meta-llama/Llama-Guard-3-11B-Vision
provider_id: null provider_id: null
provider_model_id: null provider_model_id: fireworks/llama-guard-3-11b-vision
shields: shields:
- params: null - params: null
shield_id: meta-llama/Llama-Guard-3-8B shield_id: meta-llama/Llama-Guard-3-8B

32
llama_stack/templates/together/run.yaml
View file

@ -45,37 +45,37 @@ metadata_store:
db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/together}/registry.db db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/together}/registry.db
models: models:
- metadata: {} - metadata: {}
model_id: meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo model_id: meta-llama/Llama-3.1-8B-Instruct
provider_id: null provider_id: null
provider_model_id: null provider_model_id: meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo
- metadata: {} - metadata: {}
model_id: meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo model_id: meta-llama/Llama-3.1-70B-Instruct
provider_id: null provider_id: null
provider_model_id: null provider_model_id: meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo
- metadata: {} - metadata: {}
model_id: meta-llama/Meta-Llama-3.1-405B-Instruct-Turbo model_id: meta-llama/Llama-3.1-405B-Instruct-FP8
provider_id: null provider_id: null
provider_model_id: null provider_model_id: meta-llama/Meta-Llama-3.1-405B-Instruct-Turbo
- metadata: {} - metadata: {}
model_id: meta-llama/Llama-3.2-3B-Instruct-Turbo model_id: meta-llama/Llama-3.2-3B-Instruct
provider_id: null provider_id: null
provider_model_id: null provider_model_id: meta-llama/Llama-3.2-3B-Instruct-Turbo
- metadata: {} - metadata: {}
model_id: meta-llama/Llama-3.2-11B-Vision-Instruct-Turbo model_id: meta-llama/Llama-3.2-11B-Vision-Instruct
provider_id: null provider_id: null
provider_model_id: null provider_model_id: meta-llama/Llama-3.2-11B-Vision-Instruct-Turbo
- metadata: {} - metadata: {}
model_id: meta-llama/Llama-3.2-90B-Vision-Instruct-Turbo model_id: meta-llama/Llama-3.2-90B-Vision-Instruct
provider_id: null provider_id: null
provider_model_id: null provider_model_id: meta-llama/Llama-3.2-90B-Vision-Instruct-Turbo
- metadata: {} - metadata: {}
model_id: meta-llama/Meta-Llama-Guard-3-8B model_id: meta-llama/Llama-Guard-3-8B
provider_id: null provider_id: null
provider_model_id: null provider_model_id: meta-llama/Meta-Llama-Guard-3-8B
- metadata: {} - metadata: {}
model_id: meta-llama/Llama-Guard-3-11B-Vision-Turbo model_id: meta-llama/Llama-Guard-3-11B-Vision
provider_id: null provider_id: null
provider_model_id: null provider_model_id: meta-llama/Llama-Guard-3-11B-Vision-Turbo
shields: shields:
- params: null - params: null
shield_id: meta-llama/Llama-Guard-3-1B shield_id: meta-llama/Llama-Guard-3-1B

13
llama_stack/templates/together/together.py
View file

@ -6,6 +6,8 @@
from pathlib import Path from pathlib import Path
from llama_models.sku_list import all_registered_models
from llama_stack.distribution.datatypes import ModelInput, Provider, ShieldInput from llama_stack.distribution.datatypes import ModelInput, Provider, ShieldInput
from llama_stack.providers.remote.inference.together import TogetherImplConfig from llama_stack.providers.remote.inference.together import TogetherImplConfig
from llama_stack.providers.remote.inference.together.together import MODEL_ALIASES from llama_stack.providers.remote.inference.together.together import MODEL_ALIASES
@ -28,7 +30,16 @@ def get_distribution_template() -> DistributionTemplate:
config=TogetherImplConfig.sample_run_config(), config=TogetherImplConfig.sample_run_config(),
) )
default_models = [ModelInput(model_id=m.provider_model_id) for m in MODEL_ALIASES] core_model_to_hf_repo = {
m.descriptor(): m.huggingface_repo for m in all_registered_models()
}
default_models = [
ModelInput(
model_id=core_model_to_hf_repo[m.llama_model],
provider_model_id=m.provider_model_id,
)
for m in MODEL_ALIASES
]
return DistributionTemplate( return DistributionTemplate(
name="together", name="together",