![\"Getting](\"https:\/\/www.kdnuggets.com\/wp-content\/uploads\/kdn-claude-api-python.png\"\/)
\u00a0<\/p>\n
#\u00a0<\/span>Introduction<\/h2>\n\u00a0You wish to add Claude to a Python software. Creating an account and making your first API name is easy. The official documentation can get you from zero to a working request in a couple of minutes. The following questions are often extra sensible:<\/p>\n
What does the response object comprise?
\nHow do you stream responses so customers can see output because it’s generated?
\nHow do you construction prompts and deal with responses in a manufacturing software?<\/p>\n
The Claude Python SDK takes care of a lot of the underlying API interplay. It gives typed response objects, built-in retry dealing with, and a easy interface for working with the Messages API.<\/p>\n
This text walks you thru setup, your first API name, studying the response, system prompts, and streaming. By the tip, you will have a working basis.<\/p>\n
\u00a0<\/p>\n
#\u00a0<\/span>Stipulations and Set up<\/h2>\n\u00a0You want Python 3.9 or greater, a free Claude Console account, and an API key from the Console’s Settings > API Keys web page. You possibly can add $5 in credit and work via the whole lot on this article.<\/p>\n
With these in place, set up the SDK:<\/p>\n
\u00a0<\/p>\n
By no means hardcode your API key in supply information. Retailer it as an surroundings variable as a substitute:<\/p>\n
\nexport ANTHROPIC_API_KEY=”YOUR-API-KEY-HERE”\n<\/div>\n\u00a0<\/p>\n
Or add it to a .env file on the venture root in case you’re utilizing python-dotenv. The SDK reads the ANTHROPIC_API_KEY out of your surroundings, so that you needn’t move it anyplace in your code.<\/p>\n
\u00a0<\/p>\n
#\u00a0<\/span>Making Your First API Name<\/h2>\n\u00a0The entry level for each interplay is shopper.messages.create(). Let’s ask Claude to clarify what a context window is, one thing you will really want to know as you utilize the API.<\/p>\n
You move three issues: the mannequin ID, a max_tokens restrict, and a messages checklist. The messages checklist is at all times an inventory of dicts, every with a “position” and “content material” key.<\/p>\n
\nimport anthropic
\n
\nshopper = anthropic.Anthropic()
\n
\nresponse = shopper.messages.create(
\n mannequin=”claude-sonnet-5″,
\n max_tokens=256,
\n messages=[
\n {
\n “role”: “user”,
\n “content”: “In one sentence, what is a context window?”
\n }
\n ]
\n)
\n
\nprint(response.content material[0].textual content)\n<\/div>\n\u00a0<\/p>\n
The mannequin discipline takes the precise mannequin ID string. max_tokens is a tough ceiling on what number of output tokens Claude will produce; the response stops there even when the thought is not full, so set it excessive sufficient for open-ended requests. The messages checklist should at all times begin with a “consumer” flip.<\/p>\n
Pattern output:<\/p>\n
\nA context window is the utmost quantity of textual content (measured in tokens) {that a} language
\nmannequin can course of and think about at one time, encompassing each your enter and its output.\n<\/div>\n\u00a0<\/p>\n
#\u00a0<\/span>Understanding the Response Object<\/h2>\n\u00a0The response from messages.create() is a typed Message object. It is price inspecting the total construction earlier than constructing something on prime of it.<\/p>\n
Substitute the print line within the earlier instance with:<\/p>\n
\u00a0<\/p>\n
Working that provides you the total object:<\/p>\n
\nMessage(
\n id=’msg_01XFDUDYJgAACzvnptvVoYEL’,
\n sort=”message”,
\n position=”assistant”,
\n content material=[TextBlock(text=”A context window is…”, type=”text”)],
\n mannequin=”claude-sonnet-5″,
\n stop_reason=’end_turn’,
\n stop_sequence=None,
\n utilization=Utilization(input_tokens=19, output_tokens=42)
\n)\n<\/div>\n\u00a0<\/p>\n
A number of fields right here matter greater than they first seem. stop_reason tells you why Claude stopped producing. end_turn means Claude completed by itself phrases. In the event you see max_tokens, the response was minimize off by your restrict, and you could want to lift it or rethink the immediate.<\/p>\n
The utilization discipline tracks each enter and output tokens for the request. That is how Anthropic calculates billing, and it is also the way you detect when a immediate is creeping too near the mannequin’s context restrict. content material is an inventory \u2014 in normal textual content responses it at all times has one merchandise, a TextBlock \u2014 so response.content material[0].textual content is the idiomatic strategy to pull the textual content out.<\/p>\n
\u00a0<\/p>\n
#\u00a0<\/span>Utilizing System Prompts<\/h2>\n\u00a0A system immediate permits you to give Claude a persistent position, set constraints, or present context that ought to apply throughout your entire dialog. You move it as a top-level system parameter \u2014 separate from the messages checklist, not as a message itself.<\/p>\n
Right here we configure Claude to behave as a code reviewer who solely responds in Python and avoids normal explanations:<\/p>\n
\nimport anthropic
\n
\nshopper = anthropic.Anthropic()
\n
\nresponse = shopper.messages.create(
\n mannequin=”claude-sonnet-5″,
\n max_tokens=512,
\n system=(
\n “You’re a Python code reviewer. “
\n “Reply solely with corrected or improved Python code. “
\n “Don’t clarify adjustments except the consumer explicitly asks.”
\n ),
\n messages=[
\n {
\n “role”: “user”,
\n “content”: (
\n “def get_user(id):n”
\n ” db = connect()n”
\n ” return db.query(‘SELECT * FROM users WHERE id=’ + id)”
\n )
\n }
\n ]
\n)
\n
\nprint(response.content material[0].textual content)\n<\/div>\n\u00a0<\/p>\n
The system immediate sits above the dialog in Claude’s context. It carries the identical authority all through all turns, so position directions, formatting guidelines, and area constraints you set right here persist with out you repeating them in each message.<\/p>\n
\u00a0<\/p>\n
#\u00a0<\/span>Streaming Responses<\/h2>\n\u00a0For requests the place Claude could take a couple of seconds to reply, streaming permits you to show textual content because it arrives as a substitute of ready for the total response. The SDK exposes this via shopper.messages.stream(), used as a context supervisor.<\/p>\n
The text_stream iterator yields particular person textual content chunks in actual time. Every chunk is a string fragment, not a full sentence. You move finish=”” and flush=True to print() so output seems constantly somewhat than buffering:<\/p>\n
\nimport anthropic
\n
\nshopper = anthropic.Anthropic()
\n
\nwith shopper.messages.stream(
\n mannequin=”claude-sonnet-5″,
\n max_tokens=512,
\n messages=[
\n {
\n “role”: “user”,
\n “content”: “Walk me through what happens when a Python list grows beyond its initial capacity.”
\n }
\n ]
\n) as stream:
\n for chunk in stream.text_stream:
\n print(chunk, finish=””, flush=True)
\n
\nprint() # newline after stream ends\n<\/div>\n\u00a0<\/p>\n
The context supervisor ensures the HTTP connection is closed cleanly when the block exits, even when an exception is raised mid-stream. In the event you want the entire Message object after streaming \u2014 together with token utilization counts \u2014 name stream.get_final_message() earlier than the block closes.<\/p>\n
Pattern output:<\/p>\n
\nHow do you stream responses so customers can see output because it’s generated?
\nHow do you construction prompts and deal with responses in a manufacturing software?<\/p>\n
\u00a0You want Python 3.9 or greater, a free Claude Console account, and an API key from the Console’s Settings > API Keys web page. You possibly can add $5 in credit and work via the whole lot on this article.<\/p>\n
With these in place, set up the SDK:<\/p>\n
\u00a0<\/p>\n
By no means hardcode your API key in supply information. Retailer it as an surroundings variable as a substitute:<\/p>\n
\u00a0<\/p>\n
Or add it to a .env file on the venture root in case you’re utilizing python-dotenv. The SDK reads the ANTHROPIC_API_KEY out of your surroundings, so that you needn’t move it anyplace in your code.<\/p>\n
\u00a0<\/p>\n
#\u00a0<\/span>Making Your First API Name<\/h2>\n\u00a0The entry level for each interplay is shopper.messages.create(). Let’s ask Claude to clarify what a context window is, one thing you will really want to know as you utilize the API.<\/p>\n
You move three issues: the mannequin ID, a max_tokens restrict, and a messages checklist. The messages checklist is at all times an inventory of dicts, every with a “position” and “content material” key.<\/p>\n
\nimport anthropic
\n
\nshopper = anthropic.Anthropic()
\n
\nresponse = shopper.messages.create(
\n mannequin=”claude-sonnet-5″,
\n max_tokens=256,
\n messages=[
\n {
\n “role”: “user”,
\n “content”: “In one sentence, what is a context window?”
\n }
\n ]
\n)
\n
\nprint(response.content material[0].textual content)\n<\/div>\n\u00a0<\/p>\n
The mannequin discipline takes the precise mannequin ID string. max_tokens is a tough ceiling on what number of output tokens Claude will produce; the response stops there even when the thought is not full, so set it excessive sufficient for open-ended requests. The messages checklist should at all times begin with a “consumer” flip.<\/p>\n
Pattern output:<\/p>\n
\nA context window is the utmost quantity of textual content (measured in tokens) {that a} language
\nmannequin can course of and think about at one time, encompassing each your enter and its output.\n<\/div>\n\u00a0<\/p>\n
#\u00a0<\/span>Understanding the Response Object<\/h2>\n\u00a0The response from messages.create() is a typed Message object. It is price inspecting the total construction earlier than constructing something on prime of it.<\/p>\n
Substitute the print line within the earlier instance with:<\/p>\n
\u00a0<\/p>\n
Working that provides you the total object:<\/p>\n
\nMessage(
\n id=’msg_01XFDUDYJgAACzvnptvVoYEL’,
\n sort=”message”,
\n position=”assistant”,
\n content material=[TextBlock(text=”A context window is…”, type=”text”)],
\n mannequin=”claude-sonnet-5″,
\n stop_reason=’end_turn’,
\n stop_sequence=None,
\n utilization=Utilization(input_tokens=19, output_tokens=42)
\n)\n<\/div>\n\u00a0<\/p>\n
A number of fields right here matter greater than they first seem. stop_reason tells you why Claude stopped producing. end_turn means Claude completed by itself phrases. In the event you see max_tokens, the response was minimize off by your restrict, and you could want to lift it or rethink the immediate.<\/p>\n
The utilization discipline tracks each enter and output tokens for the request. That is how Anthropic calculates billing, and it is also the way you detect when a immediate is creeping too near the mannequin’s context restrict. content material is an inventory \u2014 in normal textual content responses it at all times has one merchandise, a TextBlock \u2014 so response.content material[0].textual content is the idiomatic strategy to pull the textual content out.<\/p>\n
\u00a0<\/p>\n
#\u00a0<\/span>Utilizing System Prompts<\/h2>\n\u00a0A system immediate permits you to give Claude a persistent position, set constraints, or present context that ought to apply throughout your entire dialog. You move it as a top-level system parameter \u2014 separate from the messages checklist, not as a message itself.<\/p>\n
Right here we configure Claude to behave as a code reviewer who solely responds in Python and avoids normal explanations:<\/p>\n
\nimport anthropic
\n
\nshopper = anthropic.Anthropic()
\n
\nresponse = shopper.messages.create(
\n mannequin=”claude-sonnet-5″,
\n max_tokens=512,
\n system=(
\n “You’re a Python code reviewer. “
\n “Reply solely with corrected or improved Python code. “
\n “Don’t clarify adjustments except the consumer explicitly asks.”
\n ),
\n messages=[
\n {
\n “role”: “user”,
\n “content”: (
\n “def get_user(id):n”
\n ” db = connect()n”
\n ” return db.query(‘SELECT * FROM users WHERE id=’ + id)”
\n )
\n }
\n ]
\n)
\n
\nprint(response.content material[0].textual content)\n<\/div>\n\u00a0<\/p>\n
The system immediate sits above the dialog in Claude’s context. It carries the identical authority all through all turns, so position directions, formatting guidelines, and area constraints you set right here persist with out you repeating them in each message.<\/p>\n
\u00a0<\/p>\n
#\u00a0<\/span>Streaming Responses<\/h2>\n\u00a0For requests the place Claude could take a couple of seconds to reply, streaming permits you to show textual content because it arrives as a substitute of ready for the total response. The SDK exposes this via shopper.messages.stream(), used as a context supervisor.<\/p>\n
The text_stream iterator yields particular person textual content chunks in actual time. Every chunk is a string fragment, not a full sentence. You move finish=”” and flush=True to print() so output seems constantly somewhat than buffering:<\/p>\n
\nimport anthropic
\n
\nshopper = anthropic.Anthropic()
\n
\nwith shopper.messages.stream(
\n mannequin=”claude-sonnet-5″,
\n max_tokens=512,
\n messages=[
\n {
\n “role”: “user”,
\n “content”: “Walk me through what happens when a Python list grows beyond its initial capacity.”
\n }
\n ]
\n) as stream:
\n for chunk in stream.text_stream:
\n print(chunk, finish=””, flush=True)
\n
\nprint() # newline after stream ends\n<\/div>\n\u00a0<\/p>\n
The context supervisor ensures the HTTP connection is closed cleanly when the block exits, even when an exception is raised mid-stream. In the event you want the entire Message object after streaming \u2014 together with token utilization counts \u2014 name stream.get_final_message() earlier than the block closes.<\/p>\n
Pattern output:<\/p>\n
\n
\nshopper = anthropic.Anthropic()
\n
\nresponse = shopper.messages.create(
\n mannequin=”claude-sonnet-5″,
\n max_tokens=256,
\n messages=[
\n {
\n “role”: “user”,
\n “content”: “In one sentence, what is a context window?”
\n }
\n ]
\n)
\n
\nprint(response.content material[0].textual content)\n<\/div>\n
\u00a0<\/p>\n
The mannequin discipline takes the precise mannequin ID string. max_tokens is a tough ceiling on what number of output tokens Claude will produce; the response stops there even when the thought is not full, so set it excessive sufficient for open-ended requests. The messages checklist should at all times begin with a “consumer” flip.<\/p>\n
Pattern output:<\/p>\n
\nmannequin can course of and think about at one time, encompassing each your enter and its output.\n<\/div>\n
\u00a0<\/p>\n
#\u00a0<\/span>Understanding the Response Object<\/h2>\n\u00a0The response from messages.create() is a typed Message object. It is price inspecting the total construction earlier than constructing something on prime of it.<\/p>\n
Substitute the print line within the earlier instance with:<\/p>\n
\u00a0<\/p>\n
Working that provides you the total object:<\/p>\n
\nMessage(
\n id=’msg_01XFDUDYJgAACzvnptvVoYEL’,
\n sort=”message”,
\n position=”assistant”,
\n content material=[TextBlock(text=”A context window is…”, type=”text”)],
\n mannequin=”claude-sonnet-5″,
\n stop_reason=’end_turn’,
\n stop_sequence=None,
\n utilization=Utilization(input_tokens=19, output_tokens=42)
\n)\n<\/div>\n\u00a0<\/p>\n
A number of fields right here matter greater than they first seem. stop_reason tells you why Claude stopped producing. end_turn means Claude completed by itself phrases. In the event you see max_tokens, the response was minimize off by your restrict, and you could want to lift it or rethink the immediate.<\/p>\n
The utilization discipline tracks each enter and output tokens for the request. That is how Anthropic calculates billing, and it is also the way you detect when a immediate is creeping too near the mannequin’s context restrict. content material is an inventory \u2014 in normal textual content responses it at all times has one merchandise, a TextBlock \u2014 so response.content material[0].textual content is the idiomatic strategy to pull the textual content out.<\/p>\n
\u00a0<\/p>\n
#\u00a0<\/span>Utilizing System Prompts<\/h2>\n\u00a0A system immediate permits you to give Claude a persistent position, set constraints, or present context that ought to apply throughout your entire dialog. You move it as a top-level system parameter \u2014 separate from the messages checklist, not as a message itself.<\/p>\n
Right here we configure Claude to behave as a code reviewer who solely responds in Python and avoids normal explanations:<\/p>\n
\nimport anthropic
\n
\nshopper = anthropic.Anthropic()
\n
\nresponse = shopper.messages.create(
\n mannequin=”claude-sonnet-5″,
\n max_tokens=512,
\n system=(
\n “You’re a Python code reviewer. “
\n “Reply solely with corrected or improved Python code. “
\n “Don’t clarify adjustments except the consumer explicitly asks.”
\n ),
\n messages=[
\n {
\n “role”: “user”,
\n “content”: (
\n “def get_user(id):n”
\n ” db = connect()n”
\n ” return db.query(‘SELECT * FROM users WHERE id=’ + id)”
\n )
\n }
\n ]
\n)
\n
\nprint(response.content material[0].textual content)\n<\/div>\n\u00a0<\/p>\n
The system immediate sits above the dialog in Claude’s context. It carries the identical authority all through all turns, so position directions, formatting guidelines, and area constraints you set right here persist with out you repeating them in each message.<\/p>\n
\u00a0<\/p>\n
#\u00a0<\/span>Streaming Responses<\/h2>\n\u00a0For requests the place Claude could take a couple of seconds to reply, streaming permits you to show textual content because it arrives as a substitute of ready for the total response. The SDK exposes this via shopper.messages.stream(), used as a context supervisor.<\/p>\n
The text_stream iterator yields particular person textual content chunks in actual time. Every chunk is a string fragment, not a full sentence. You move finish=”” and flush=True to print() so output seems constantly somewhat than buffering:<\/p>\n
\nimport anthropic
\n
\nshopper = anthropic.Anthropic()
\n
\nwith shopper.messages.stream(
\n mannequin=”claude-sonnet-5″,
\n max_tokens=512,
\n messages=[
\n {
\n “role”: “user”,
\n “content”: “Walk me through what happens when a Python list grows beyond its initial capacity.”
\n }
\n ]
\n) as stream:
\n for chunk in stream.text_stream:
\n print(chunk, finish=””, flush=True)
\n
\nprint() # newline after stream ends\n<\/div>\n\u00a0<\/p>\n
The context supervisor ensures the HTTP connection is closed cleanly when the block exits, even when an exception is raised mid-stream. In the event you want the entire Message object after streaming \u2014 together with token utilization counts \u2014 name stream.get_final_message() earlier than the block closes.<\/p>\n
Pattern output:<\/p>\n
\n id=’msg_01XFDUDYJgAACzvnptvVoYEL’,
\n sort=”message”,
\n position=”assistant”,
\n content material=[TextBlock(text=”A context window is…”, type=”text”)],
\n mannequin=”claude-sonnet-5″,
\n stop_reason=’end_turn’,
\n stop_sequence=None,
\n utilization=Utilization(input_tokens=19, output_tokens=42)
\n)\n<\/div>\n
\u00a0<\/p>\n
A number of fields right here matter greater than they first seem. stop_reason tells you why Claude stopped producing. end_turn means Claude completed by itself phrases. In the event you see max_tokens, the response was minimize off by your restrict, and you could want to lift it or rethink the immediate.<\/p>\n
The utilization discipline tracks each enter and output tokens for the request. That is how Anthropic calculates billing, and it is also the way you detect when a immediate is creeping too near the mannequin’s context restrict. content material is an inventory \u2014 in normal textual content responses it at all times has one merchandise, a TextBlock \u2014 so response.content material[0].textual content is the idiomatic strategy to pull the textual content out.<\/p>\n
\u00a0<\/p>\n
#\u00a0<\/span>Utilizing System Prompts<\/h2>\n\u00a0A system immediate permits you to give Claude a persistent position, set constraints, or present context that ought to apply throughout your entire dialog. You move it as a top-level system parameter \u2014 separate from the messages checklist, not as a message itself.<\/p>\n
Right here we configure Claude to behave as a code reviewer who solely responds in Python and avoids normal explanations:<\/p>\n
\nimport anthropic
\n
\nshopper = anthropic.Anthropic()
\n
\nresponse = shopper.messages.create(
\n mannequin=”claude-sonnet-5″,
\n max_tokens=512,
\n system=(
\n “You’re a Python code reviewer. “
\n “Reply solely with corrected or improved Python code. “
\n “Don’t clarify adjustments except the consumer explicitly asks.”
\n ),
\n messages=[
\n {
\n “role”: “user”,
\n “content”: (
\n “def get_user(id):n”
\n ” db = connect()n”
\n ” return db.query(‘SELECT * FROM users WHERE id=’ + id)”
\n )
\n }
\n ]
\n)
\n
\nprint(response.content material[0].textual content)\n<\/div>\n\u00a0<\/p>\n
The system immediate sits above the dialog in Claude’s context. It carries the identical authority all through all turns, so position directions, formatting guidelines, and area constraints you set right here persist with out you repeating them in each message.<\/p>\n
\u00a0<\/p>\n
#\u00a0<\/span>Streaming Responses<\/h2>\n\u00a0For requests the place Claude could take a couple of seconds to reply, streaming permits you to show textual content because it arrives as a substitute of ready for the total response. The SDK exposes this via shopper.messages.stream(), used as a context supervisor.<\/p>\n
The text_stream iterator yields particular person textual content chunks in actual time. Every chunk is a string fragment, not a full sentence. You move finish=”” and flush=True to print() so output seems constantly somewhat than buffering:<\/p>\n
\nimport anthropic
\n
\nshopper = anthropic.Anthropic()
\n
\nwith shopper.messages.stream(
\n mannequin=”claude-sonnet-5″,
\n max_tokens=512,
\n messages=[
\n {
\n “role”: “user”,
\n “content”: “Walk me through what happens when a Python list grows beyond its initial capacity.”
\n }
\n ]
\n) as stream:
\n for chunk in stream.text_stream:
\n print(chunk, finish=””, flush=True)
\n
\nprint() # newline after stream ends\n<\/div>\n\u00a0<\/p>\n
The context supervisor ensures the HTTP connection is closed cleanly when the block exits, even when an exception is raised mid-stream. In the event you want the entire Message object after streaming \u2014 together with token utilization counts \u2014 name stream.get_final_message() earlier than the block closes.<\/p>\n
Pattern output:<\/p>\n
\n
\nshopper = anthropic.Anthropic()
\n
\nresponse = shopper.messages.create(
\n mannequin=”claude-sonnet-5″,
\n max_tokens=512,
\n system=(
\n “You’re a Python code reviewer. “
\n “Reply solely with corrected or improved Python code. “
\n “Don’t clarify adjustments except the consumer explicitly asks.”
\n ),
\n messages=[
\n {
\n “role”: “user”,
\n “content”: (
\n “def get_user(id):n”
\n ” db = connect()n”
\n ” return db.query(‘SELECT * FROM users WHERE id=’ + id)”
\n )
\n }
\n ]
\n)
\n
\nprint(response.content material[0].textual content)\n<\/div>\n
\u00a0<\/p>\n
The system immediate sits above the dialog in Claude’s context. It carries the identical authority all through all turns, so position directions, formatting guidelines, and area constraints you set right here persist with out you repeating them in each message.<\/p>\n
\u00a0<\/p>\n
#\u00a0<\/span>Streaming Responses<\/h2>\n\u00a0For requests the place Claude could take a couple of seconds to reply, streaming permits you to show textual content because it arrives as a substitute of ready for the total response. The SDK exposes this via shopper.messages.stream(), used as a context supervisor.<\/p>\n
The text_stream iterator yields particular person textual content chunks in actual time. Every chunk is a string fragment, not a full sentence. You move finish=”” and flush=True to print() so output seems constantly somewhat than buffering:<\/p>\n
\nimport anthropic
\n
\nshopper = anthropic.Anthropic()
\n
\nwith shopper.messages.stream(
\n mannequin=”claude-sonnet-5″,
\n max_tokens=512,
\n messages=[
\n {
\n “role”: “user”,
\n “content”: “Walk me through what happens when a Python list grows beyond its initial capacity.”
\n }
\n ]
\n) as stream:
\n for chunk in stream.text_stream:
\n print(chunk, finish=””, flush=True)
\n
\nprint() # newline after stream ends\n<\/div>\n\u00a0<\/p>\n
The context supervisor ensures the HTTP connection is closed cleanly when the block exits, even when an exception is raised mid-stream. In the event you want the entire Message object after streaming \u2014 together with token utilization counts \u2014 name stream.get_final_message() earlier than the block closes.<\/p>\n
Pattern output:<\/p>\n
\n
\nshopper = anthropic.Anthropic()
\n
\nwith shopper.messages.stream(
\n mannequin=”claude-sonnet-5″,
\n max_tokens=512,
\n messages=[
\n {
\n “role”: “user”,
\n “content”: “Walk me through what happens when a Python list grows beyond its initial capacity.”
\n }
\n ]
\n) as stream:
\n for chunk in stream.text_stream:
\n print(chunk, finish=””, flush=True)
\n
\nprint() # newline after stream ends\n<\/div>\n
\u00a0<\/p>\n
The context supervisor ensures the HTTP connection is closed cleanly when the block exits, even when an exception is raised mid-stream. In the event you want the entire Message object after streaming \u2014 together with token utilization counts \u2014 name stream.get_final_message() earlier than the block closes.<\/p>\n
Pattern output:<\/p>\n