File size: 5,161 Bytes
b8755c7 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 |
import json
from uuid import uuid4
from open_webui.utils.misc import (
openai_chat_chunk_message_template,
openai_chat_completion_message_template,
)
def convert_ollama_tool_call_to_openai(tool_calls: dict) -> dict:
openai_tool_calls = []
for tool_call in tool_calls:
openai_tool_call = {
"index": tool_call.get("index", 0),
"id": tool_call.get("id", f"call_{str(uuid4())}"),
"type": "function",
"function": {
"name": tool_call.get("function", {}).get("name", ""),
"arguments": json.dumps(
tool_call.get("function", {}).get("arguments", {})
),
},
}
openai_tool_calls.append(openai_tool_call)
return openai_tool_calls
def convert_response_ollama_to_openai(ollama_response: dict) -> dict:
model = ollama_response.get("model", "ollama")
message_content = ollama_response.get("message", {}).get("content", "")
tool_calls = ollama_response.get("message", {}).get("tool_calls", None)
openai_tool_calls = None
if tool_calls:
openai_tool_calls = convert_ollama_tool_call_to_openai(tool_calls)
data = ollama_response
usage = {
"response_token/s": (
round(
(
(
data.get("eval_count", 0)
/ ((data.get("eval_duration", 0) / 10_000_000))
)
* 100
),
2,
)
if data.get("eval_duration", 0) > 0
else "N/A"
),
"prompt_token/s": (
round(
(
(
data.get("prompt_eval_count", 0)
/ ((data.get("prompt_eval_duration", 0) / 10_000_000))
)
* 100
),
2,
)
if data.get("prompt_eval_duration", 0) > 0
else "N/A"
),
"total_duration": data.get("total_duration", 0),
"load_duration": data.get("load_duration", 0),
"prompt_eval_count": data.get("prompt_eval_count", 0),
"prompt_eval_duration": data.get("prompt_eval_duration", 0),
"eval_count": data.get("eval_count", 0),
"eval_duration": data.get("eval_duration", 0),
"approximate_total": (lambda s: f"{s // 3600}h{(s % 3600) // 60}m{s % 60}s")(
(data.get("total_duration", 0) or 0) // 1_000_000_000
),
}
response = openai_chat_completion_message_template(
model, message_content, openai_tool_calls, usage
)
return response
async def convert_streaming_response_ollama_to_openai(ollama_streaming_response):
async for data in ollama_streaming_response.body_iterator:
data = json.loads(data)
model = data.get("model", "ollama")
message_content = data.get("message", {}).get("content", "")
tool_calls = data.get("message", {}).get("tool_calls", None)
openai_tool_calls = None
if tool_calls:
openai_tool_calls = convert_ollama_tool_call_to_openai(tool_calls)
done = data.get("done", False)
usage = None
if done:
usage = {
"response_token/s": (
round(
(
(
data.get("eval_count", 0)
/ ((data.get("eval_duration", 0) / 10_000_000))
)
* 100
),
2,
)
if data.get("eval_duration", 0) > 0
else "N/A"
),
"prompt_token/s": (
round(
(
(
data.get("prompt_eval_count", 0)
/ ((data.get("prompt_eval_duration", 0) / 10_000_000))
)
* 100
),
2,
)
if data.get("prompt_eval_duration", 0) > 0
else "N/A"
),
"total_duration": data.get("total_duration", 0),
"load_duration": data.get("load_duration", 0),
"prompt_eval_count": data.get("prompt_eval_count", 0),
"prompt_eval_duration": data.get("prompt_eval_duration", 0),
"eval_count": data.get("eval_count", 0),
"eval_duration": data.get("eval_duration", 0),
"approximate_total": (
lambda s: f"{s // 3600}h{(s % 3600) // 60}m{s % 60}s"
)((data.get("total_duration", 0) or 0) // 1_000_000_000),
}
data = openai_chat_chunk_message_template(
model, message_content if not done else None, openai_tool_calls, usage
)
line = f"data: {json.dumps(data)}\n\n"
yield line
yield "data: [DONE]\n\n"
|