Update the tests.

This commit is contained in:
Nicolas Patry 2025-03-11 13:19:38 +01:00
parent 03fe626a95
commit ef4e2685d8
No known key found for this signature in database
GPG Key ID: 4242CEF24CB6DBF9
4 changed files with 342 additions and 1256 deletions

View File

@ -5,20 +5,20 @@
"index": 0,
"logprobs": null,
"message": {
"content": "I am a helpful assistant!",
"content": "I'm an artificial intelligence model known as a large language model (LLM) or conversational AI",
"role": "assistant",
"tool_calls": null
}
}
],
"created": 1741263686,
"created": 1741693957,
"id": "",
"model": "meta-llama/Llama-3.1-8B-Instruct",
"object": "chat.completion",
"system_fingerprint": "3.1.2-dev0-native",
"usage": {
"completion_tokens": 23,
"prompt_tokens": 494,
"total_tokens": 517
"completion_tokens": 12,
"prompt_tokens": 53,
"total_tokens": 65
}
}

View File

@ -1,24 +1,4 @@
[
{
"choices": [
{
"delta": {
"content": "",
"role": "assistant",
"tool_calls": null
},
"finish_reason": null,
"index": 0,
"logprobs": null
}
],
"created": 1741364571,
"id": "",
"model": "meta-llama/Llama-3.1-8B-Instruct",
"object": "chat.completion.chunk",
"system_fingerprint": "3.1.2-dev0-native",
"usage": null
},
{
"choices": [
{
@ -32,7 +12,7 @@
"logprobs": null
}
],
"created": 1741364571,
"created": 1741694017,
"id": "",
"model": "meta-llama/Llama-3.1-8B-Instruct",
"object": "chat.completion.chunk",
@ -43,7 +23,7 @@
"choices": [
{
"delta": {
"content": " am",
"content": "'m",
"role": "assistant",
"tool_calls": null
},
@ -52,7 +32,127 @@
"logprobs": null
}
],
"created": 1741364571,
"created": 1741694017,
"id": "",
"model": "meta-llama/Llama-3.1-8B-Instruct",
"object": "chat.completion.chunk",
"system_fingerprint": "3.1.2-dev0-native",
"usage": null
},
{
"choices": [
{
"delta": {
"content": " an",
"role": "assistant",
"tool_calls": null
},
"finish_reason": null,
"index": 0,
"logprobs": null
}
],
"created": 1741694017,
"id": "",
"model": "meta-llama/Llama-3.1-8B-Instruct",
"object": "chat.completion.chunk",
"system_fingerprint": "3.1.2-dev0-native",
"usage": null
},
{
"choices": [
{
"delta": {
"content": " artificial",
"role": "assistant",
"tool_calls": null
},
"finish_reason": null,
"index": 0,
"logprobs": null
}
],
"created": 1741694017,
"id": "",
"model": "meta-llama/Llama-3.1-8B-Instruct",
"object": "chat.completion.chunk",
"system_fingerprint": "3.1.2-dev0-native",
"usage": null
},
{
"choices": [
{
"delta": {
"content": " intelligence",
"role": "assistant",
"tool_calls": null
},
"finish_reason": null,
"index": 0,
"logprobs": null
}
],
"created": 1741694017,
"id": "",
"model": "meta-llama/Llama-3.1-8B-Instruct",
"object": "chat.completion.chunk",
"system_fingerprint": "3.1.2-dev0-native",
"usage": null
},
{
"choices": [
{
"delta": {
"content": " model",
"role": "assistant",
"tool_calls": null
},
"finish_reason": null,
"index": 0,
"logprobs": null
}
],
"created": 1741694017,
"id": "",
"model": "meta-llama/Llama-3.1-8B-Instruct",
"object": "chat.completion.chunk",
"system_fingerprint": "3.1.2-dev0-native",
"usage": null
},
{
"choices": [
{
"delta": {
"content": " known",
"role": "assistant",
"tool_calls": null
},
"finish_reason": null,
"index": 0,
"logprobs": null
}
],
"created": 1741694017,
"id": "",
"model": "meta-llama/Llama-3.1-8B-Instruct",
"object": "chat.completion.chunk",
"system_fingerprint": "3.1.2-dev0-native",
"usage": null
},
{
"choices": [
{
"delta": {
"content": " as",
"role": "assistant",
"tool_calls": null
},
"finish_reason": null,
"index": 0,
"logprobs": null
}
],
"created": 1741694017,
"id": "",
"model": "meta-llama/Llama-3.1-8B-Instruct",
"object": "chat.completion.chunk",
@ -72,7 +172,7 @@
"logprobs": null
}
],
"created": 1741364571,
"created": 1741694017,
"id": "",
"model": "meta-llama/Llama-3.1-8B-Instruct",
"object": "chat.completion.chunk",
@ -83,7 +183,7 @@
"choices": [
{
"delta": {
"content": " helpful",
"content": " large",
"role": "assistant",
"tool_calls": null
},
@ -92,7 +192,7 @@
"logprobs": null
}
],
"created": 1741364571,
"created": 1741694017,
"id": "",
"model": "meta-llama/Llama-3.1-8B-Instruct",
"object": "chat.completion.chunk",
@ -103,7 +203,7 @@
"choices": [
{
"delta": {
"content": " assistant",
"content": " language",
"role": "assistant",
"tool_calls": null
},
@ -112,7 +212,7 @@
"logprobs": null
}
],
"created": 1741364571,
"created": 1741694017,
"id": "",
"model": "meta-llama/Llama-3.1-8B-Instruct",
"object": "chat.completion.chunk",
@ -123,7 +223,7 @@
"choices": [
{
"delta": {
"content": "!",
"content": " model",
"role": "assistant",
"tool_calls": null
},
@ -132,7 +232,167 @@
"logprobs": null
}
],
"created": 1741364571,
"created": 1741694017,
"id": "",
"model": "meta-llama/Llama-3.1-8B-Instruct",
"object": "chat.completion.chunk",
"system_fingerprint": "3.1.2-dev0-native",
"usage": null
},
{
"choices": [
{
"delta": {
"content": " (",
"role": "assistant",
"tool_calls": null
},
"finish_reason": null,
"index": 0,
"logprobs": null
}
],
"created": 1741694017,
"id": "",
"model": "meta-llama/Llama-3.1-8B-Instruct",
"object": "chat.completion.chunk",
"system_fingerprint": "3.1.2-dev0-native",
"usage": null
},
{
"choices": [
{
"delta": {
"content": "LL",
"role": "assistant",
"tool_calls": null
},
"finish_reason": null,
"index": 0,
"logprobs": null
}
],
"created": 1741694017,
"id": "",
"model": "meta-llama/Llama-3.1-8B-Instruct",
"object": "chat.completion.chunk",
"system_fingerprint": "3.1.2-dev0-native",
"usage": null
},
{
"choices": [
{
"delta": {
"content": "M",
"role": "assistant",
"tool_calls": null
},
"finish_reason": null,
"index": 0,
"logprobs": null
}
],
"created": 1741694017,
"id": "",
"model": "meta-llama/Llama-3.1-8B-Instruct",
"object": "chat.completion.chunk",
"system_fingerprint": "3.1.2-dev0-native",
"usage": null
},
{
"choices": [
{
"delta": {
"content": ")",
"role": "assistant",
"tool_calls": null
},
"finish_reason": null,
"index": 0,
"logprobs": null
}
],
"created": 1741694017,
"id": "",
"model": "meta-llama/Llama-3.1-8B-Instruct",
"object": "chat.completion.chunk",
"system_fingerprint": "3.1.2-dev0-native",
"usage": null
},
{
"choices": [
{
"delta": {
"content": " or",
"role": "assistant",
"tool_calls": null
},
"finish_reason": null,
"index": 0,
"logprobs": null
}
],
"created": 1741694017,
"id": "",
"model": "meta-llama/Llama-3.1-8B-Instruct",
"object": "chat.completion.chunk",
"system_fingerprint": "3.1.2-dev0-native",
"usage": null
},
{
"choices": [
{
"delta": {
"content": " convers",
"role": "assistant",
"tool_calls": null
},
"finish_reason": null,
"index": 0,
"logprobs": null
}
],
"created": 1741694017,
"id": "",
"model": "meta-llama/Llama-3.1-8B-Instruct",
"object": "chat.completion.chunk",
"system_fingerprint": "3.1.2-dev0-native",
"usage": null
},
{
"choices": [
{
"delta": {
"content": "ational",
"role": "assistant",
"tool_calls": null
},
"finish_reason": null,
"index": 0,
"logprobs": null
}
],
"created": 1741694017,
"id": "",
"model": "meta-llama/Llama-3.1-8B-Instruct",
"object": "chat.completion.chunk",
"system_fingerprint": "3.1.2-dev0-native",
"usage": null
},
{
"choices": [
{
"delta": {
"content": " AI",
"role": "assistant",
"tool_calls": null
},
"finish_reason": "length",
"index": 0,
"logprobs": null
}
],
"created": 1741694017,
"id": "",
"model": "meta-llama/Llama-3.1-8B-Instruct",
"object": "chat.completion.chunk",

View File

@ -279,7 +279,7 @@ async def test_flash_llama_grammar_tools_insufficient_information_nostream(
):
client = InferenceClient(base_url=f"{flash_llama_grammar_tools.base_url}/v1")
response = client.chat_completion(
max_tokens=100,
max_tokens=20,
seed=24,
tools=tools,
tool_choice="auto",
@ -299,7 +299,10 @@ async def test_flash_llama_grammar_tools_insufficient_information_nostream(
content_generated = response.choices[0].message.content
assert response.choices[0].message.tool_calls is None
assert content_generated == "I am a helpful assistant!"
assert (
content_generated
== "I'm an artificial intelligence model known as a large language model (LLM) or conversational AI"
)
assert response == response_snapshot
@ -310,7 +313,7 @@ async def test_flash_llama_grammar_tools_insufficient_information_stream(
):
client = InferenceClient(base_url=f"{flash_llama_grammar_tools.base_url}/v1")
stream = client.chat_completion(
max_tokens=100,
max_tokens=20,
seed=24,
tools=tools,
tool_choice="auto",
@ -335,7 +338,10 @@ async def test_flash_llama_grammar_tools_insufficient_information_stream(
assert chunk.choices[0].delta.tool_calls is None
######## This is exactly the same as the non streaming case
assert content_generated == "I am a helpful assistant!"
assert (
content_generated
== "I'm an artificial intelligence model known as a large language model (LLM) or conversational AI"
)
assert chunks == response_snapshot
@ -346,7 +352,7 @@ async def test_flash_llama_grammar_tools_sea_creatures_stream_auto(
):
client = InferenceClient(base_url=f"{flash_llama_grammar_tools.base_url}/v1")
stream = client.chat_completion(
max_tokens=100,
max_tokens=20,
seed=24,
tools=tools,
tool_choice="auto",
@ -372,7 +378,7 @@ async def test_flash_llama_grammar_tools_sea_creatures_stream_auto(
assert (
content_generated
== "There was a wise old octopus named Oracle. He lived in a cozy little cave beneath the waves with his best friend, a curious seahorse named Finley. One day, Finley met a playful dolphin named Daisy, and the three became inseparable. They spent their days exploring the ocean, playing hide-and-seek, and learning about the wonders of the sea from Oracle."
== "Once upon a time, in a vibrant ocean filled with coral reefs and schools of shimmering fish,"
)
assert chunks == response_snapshot