Skip to content

Add CentML Llama 3.3 70B fastest and the cheapest on the market #132

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 3 commits into
base: main
Choose a base branch
from
Open

Add CentML Llama 3.3 70B fastest and the cheapest on the market #132

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
3593a7d
Add CentML provider of Llama 405b
anurlybayev Dec 4, 2024
2fc58c0
Add CentML to the list in README
anurlybayev Dec 4, 2024
24bd30a
Replace Llama 3.1 405B with Llama 3.3 70B
anurlybayev Jan 5, 2025
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
1 change: 1 addition & 0 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -10,6 +10,7 @@ Large Language Models (LLMs):
- Llama2 and 3 from several different providers, including
- Anyscale
- Azure
- CentML
- Cerebras
- Cloudflare
- Groq
Expand Down
16 changes: 16 additions & 0 deletions llm_benchmark_suite.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -27,6 +27,7 @@
GPT_35_TURBO_1106 = "gpt-3.5-turbo-1106"
GEMINI_1_5_PRO = "gemini-1.5-pro"
GEMINI_1_5_FLASH = "gemini-1.5-flash"
LLAMA_33_70B_INSTRUCT = "Llama-3.3-70B-Instruct"
LLAMA_31_405B_CHAT = "llama-3.1-405b-chat"
LLAMA_31_405B_CHAT_FP8 = "llama-3.1-405b-chat-fp8"
LLAMA_31_70B_CHAT = "llama-3.1-70b-chat"
Expand Down Expand Up @@ -125,6 +126,18 @@ async def run(self, pass_argv: List[str], spread: float) -> asyncio.Task:
return await llm_benchmark.run(full_argv)


class _CentmlLlm(_Llm):
"""See https://docs.centml.ai/resources/pricing"""

def __init__(self, model: str, display_model: Optional[str] = None):
super().__init__(
model,
"centml.ai/" + (display_model or model),
api_key=os.getenv("CENTML_API_KEY"),
base_url="https://api.centml.com/openai/v1",
)


class _CerebrasLlm(_Llm):
"""See https://docs.cerebras.ai/en/latest/wsc/Model-zoo/MZ-overview.html#list-of-models"""

Expand Down Expand Up @@ -368,6 +381,8 @@ def _text_models():
"meta-llama/Meta-Llama-3.1-405B-Instruct-Turbo", LLAMA_31_405B_CHAT_FP8
),
# _OvhLlm("llama-3-1-405b-instruct", LLAMA_31_405B_CHAT),
# Llama 3.3 70b
_CentmlLlm("meta-llama/Llama-3.3-70B-Instruct", LLAMA_33_70B_INSTRUCT),
# Llama 3.1 70b
_CerebrasLlm("llama3.1-70b", LLAMA_31_70B_CHAT),
_CloudflareLlm("@cf/meta/llama-3.1-70b-preview", LLAMA_31_70B_CHAT),
Expand Down Expand Up @@ -459,6 +474,7 @@ def _tools_models():
# _FireworksLlm(
# "accounts/fireworks/models/llama-v3p1-405b-instruct", LLAMA_31_405B_CHAT_FP8
# ), returns "FUNCTION" and the call as text
_CentmlLlm("meta-llama/Llama-3.3-70B-Instruct", LLAMA_33_70B_INSTRUCT),
_GroqLlm("llama-3.1-405b-reasoning", LLAMA_31_405B_CHAT_FP8),
_GroqLlm("llama-3.1-70b-versatile", LLAMA_31_70B_CHAT_FP8),
_GroqLlm("llama-3.1-8b-instant", LLAMA_31_8B_CHAT_FP8),
Expand Down