Update hf_client.py
Browse files- hf_client.py +20 -33
hf_client.py
CHANGED
@@ -2,61 +2,50 @@ import os
|
|
2 |
from huggingface_hub import InferenceClient
|
3 |
from tavily import TavilyClient
|
4 |
|
5 |
-
# ===
|
6 |
-
|
7 |
HF_TOKEN = os.getenv('HF_TOKEN')
|
8 |
-
GROQ_API_KEY = os.getenv(
|
9 |
TAVILY_API_KEY = os.getenv('TAVILY_API_KEY')
|
10 |
|
11 |
if not HF_TOKEN:
|
12 |
-
raise RuntimeError(
|
13 |
-
"HF_TOKEN environment variable is not set. Please set it to your Hugging Face API token."
|
14 |
-
)
|
15 |
-
|
16 |
-
|
17 |
-
# === Groq Adapter ===
|
18 |
|
|
|
19 |
class GroqChatClient:
|
20 |
-
def __init__(self, api_key: str
|
21 |
import openai
|
22 |
openai.api_key = api_key
|
23 |
openai.api_base = "https://api.groq.com/openai/v1"
|
24 |
-
self.
|
25 |
-
self.
|
26 |
|
27 |
class Chat:
|
28 |
-
def __init__(self,
|
29 |
-
self.completions = self.Completions(
|
30 |
|
31 |
class Completions:
|
32 |
-
def __init__(self,
|
33 |
-
self.client =
|
34 |
-
self.model = model
|
35 |
|
36 |
-
def create(self, messages, temperature=0.7, max_tokens=1024):
|
37 |
return self.client.ChatCompletion.create(
|
38 |
-
model=
|
39 |
messages=messages,
|
40 |
temperature=temperature,
|
41 |
-
max_tokens=max_tokens
|
|
|
42 |
)
|
43 |
|
44 |
-
|
45 |
-
def chat(self):
|
46 |
-
return self.Chat(self.openai, self.model)
|
47 |
-
|
48 |
-
|
49 |
-
# === Unified Client Factory ===
|
50 |
-
|
51 |
def get_inference_client(model_id: str, provider: str = "auto"):
|
52 |
"""
|
53 |
-
Returns a unified
|
54 |
-
-
|
55 |
-
-
|
56 |
"""
|
57 |
if model_id == "moonshotai/Kimi-K2-Instruct":
|
58 |
if not GROQ_API_KEY:
|
59 |
-
raise RuntimeError("GROQ_API_KEY is
|
60 |
return GroqChatClient(api_key=GROQ_API_KEY)
|
61 |
|
62 |
return InferenceClient(
|
@@ -66,9 +55,7 @@ def get_inference_client(model_id: str, provider: str = "auto"):
|
|
66 |
bill_to="huggingface"
|
67 |
)
|
68 |
|
69 |
-
|
70 |
# === Tavily Search Client ===
|
71 |
-
|
72 |
tavily_client = None
|
73 |
if TAVILY_API_KEY:
|
74 |
try:
|
|
|
2 |
from huggingface_hub import InferenceClient
|
3 |
from tavily import TavilyClient
|
4 |
|
5 |
+
# === API Keys ===
|
|
|
6 |
HF_TOKEN = os.getenv('HF_TOKEN')
|
7 |
+
GROQ_API_KEY = os.getenv('GROQ_API_KEY')
|
8 |
TAVILY_API_KEY = os.getenv('TAVILY_API_KEY')
|
9 |
|
10 |
if not HF_TOKEN:
|
11 |
+
raise RuntimeError("HF_TOKEN environment variable is not set. Please set it to your Hugging Face API token.")
|
|
|
|
|
|
|
|
|
|
|
12 |
|
13 |
+
# === GROQ-Compatible Wrapper ===
|
14 |
class GroqChatClient:
|
15 |
+
def __init__(self, api_key: str):
|
16 |
import openai
|
17 |
openai.api_key = api_key
|
18 |
openai.api_base = "https://api.groq.com/openai/v1"
|
19 |
+
self.client = openai
|
20 |
+
self.chat = self.Chat(openai)
|
21 |
|
22 |
class Chat:
|
23 |
+
def __init__(self, openai_client):
|
24 |
+
self.completions = self.Completions(openai_client)
|
25 |
|
26 |
class Completions:
|
27 |
+
def __init__(self, openai_client):
|
28 |
+
self.client = openai_client
|
|
|
29 |
|
30 |
+
def create(self, model, messages, temperature=0.7, max_tokens=1024, **kwargs):
|
31 |
return self.client.ChatCompletion.create(
|
32 |
+
model=model,
|
33 |
messages=messages,
|
34 |
temperature=temperature,
|
35 |
+
max_tokens=max_tokens,
|
36 |
+
**kwargs
|
37 |
)
|
38 |
|
39 |
+
# === Inference Client Selector ===
|
|
|
|
|
|
|
|
|
|
|
|
|
40 |
def get_inference_client(model_id: str, provider: str = "auto"):
|
41 |
"""
|
42 |
+
Returns a unified interface:
|
43 |
+
- For 'moonshotai/Kimi-K2-Instruct', uses Groq with OpenAI-compatible API
|
44 |
+
- For others, uses Hugging Face InferenceClient
|
45 |
"""
|
46 |
if model_id == "moonshotai/Kimi-K2-Instruct":
|
47 |
if not GROQ_API_KEY:
|
48 |
+
raise RuntimeError("GROQ_API_KEY is required for Groq-hosted models.")
|
49 |
return GroqChatClient(api_key=GROQ_API_KEY)
|
50 |
|
51 |
return InferenceClient(
|
|
|
55 |
bill_to="huggingface"
|
56 |
)
|
57 |
|
|
|
58 |
# === Tavily Search Client ===
|
|
|
59 |
tavily_client = None
|
60 |
if TAVILY_API_KEY:
|
61 |
try:
|