Chat completions

This is the core endpoint for all AI-powered features — summarization, extraction, analysis, drafting.

Endpoint

POST /llm/v1/chat/completions

curl -X POST https://api.case.dev/llm/v1/chat/completions \
  -H "Authorization: Bearer sk_case_YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "openai/gpt-6-sol",
    "messages": [
      {"role": "user", "content": "Summarize this deposition in 3 bullet points."}
    ]
  }'

casedev llm:v1:chat create-completion \
  --model openai/gpt-6-sol \
  --message '{role: user, content: "Summarize this deposition in 3 bullet points."}'

import Casedev from 'casedev';

const client = new Casedev({ apiKey: 'sk_case_YOUR_API_KEY' });

const response = await client.llm.v1.chat.createCompletion({
  model: 'openai/gpt-6-sol',
  messages: [
    { role: 'user', content: 'Summarize this deposition in 3 bullet points.' }
  ]
});

console.log(response.choices[0].message.content);

import casedev

client = casedev.Casedev(api_key='sk_case_YOUR_API_KEY')

response = client.llm.v1.chat.create_completion(
    model='openai/gpt-6-sol',
    messages=[
        {'role': 'user', 'content': 'Summarize this deposition in 3 bullet points.'}
    ]
)

print(response.choices[0].message.content)

resp, _ := client.Llm.V1.Chat.NewCompletion(ctx, casedev.LlmV1ChatNewCompletionParams{
	Model: casedev.F("openai/gpt-6-sol"),
	Messages: casedev.F([]casedev.LlmV1ChatNewCompletionParamsMessage{{
		Role:    casedev.F(casedev.LlmV1ChatNewCompletionParamsMessagesRoleUser),
		Content: casedev.F("Summarize this deposition in 3 bullet points."),
	}}),
})
fmt.Println(resp.Choices[0].Message.Content)

Response

{
  "id": "gen_01K972J7KV4Y0MJZ3SRTA6YYMH",
  "object": "chat.completion",
  "model": "openai/gpt-6-sol",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Here are the key points:\n\n• Witness testified that...\n• Documents reviewed include...\n• Timeline established from..."
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 245,
    "completion_tokens": 87,
    "total_tokens": 332,
    "cost": 0.000105
  }
}

Parameters

Required

Parameter	Type	Description
`messages`	array	The conversation. Each message has a `role` and `content`.

Optional

Parameter	Type	Default	Description
`model`	string	`casemark/core-large`	Which model to use. Browse all models →
`max_tokens`	number	4096	Maximum tokens to generate
`temperature`	number	1	Randomness (0-2). Use 0 for factual tasks.
`stream`	boolean	false	Stream response token-by-token
`stop`	array	null	Stop generation when these strings appear

Messages

Each message in the messages array:

Field	Type	Description
`role`	string	`system`, `user`, or `assistant`
`content`	string	The message text

System prompts

Set the AI’s behavior with a system message:

casedev llm:v1:chat create-completion \
  --model openai/gpt-6-sol \
  --message '{role: system, content: "You are a legal assistant. Be concise. Cite case law when relevant."}' \
  --message '{role: user, content: "What are the elements of negligence?"}'

const response = await client.llm.v1.chat.createCompletion({
  model: 'openai/gpt-6-sol',
  messages: [
    {
      role: 'system',
      content: 'You are a legal assistant. Be concise. Cite case law when relevant.'
    },
    {
      role: 'user',
      content: 'What are the elements of negligence?'
    }
  ]
});

response = client.llm.v1.chat.create_completion(
    model='openai/gpt-6-sol',
    messages=[
        {
            'role': 'system',
            'content': 'You are a legal assistant. Be concise. Cite case law when relevant.'
        },
        {
            'role': 'user',
            'content': 'What are the elements of negligence?'
        }
    ]
)

resp, _ := client.Llm.V1.Chat.NewCompletion(ctx, casedev.LlmV1ChatNewCompletionParams{
	Model: casedev.F("openai/gpt-6-sol"),
	Messages: casedev.F([]casedev.LlmV1ChatNewCompletionParamsMessage{
		{
			Role:    casedev.F(casedev.LlmV1ChatNewCompletionParamsMessagesRoleSystem),
			Content: casedev.F("You are a legal assistant. Be concise. Cite case law when relevant."),
		},
		{
			Role:    casedev.F(casedev.LlmV1ChatNewCompletionParamsMessagesRoleUser),
			Content: casedev.F("What are the elements of negligence?"),
		},
	}),
})
fmt.Println(resp.Choices[0].Message.Content)

Multi-turn conversations

Include previous messages to maintain context:

casedev llm:v1:chat create-completion \
  --model openai/gpt-6-sol \
  --message '{role: user, content: "What is a deposition?"}' \
  --message '{role: assistant, content: "A deposition is sworn testimony taken outside of court..."}' \
  --message '{role: user, content: "How long do they typically last?"}'

const response = await client.llm.v1.chat.createCompletion({
  model: 'openai/gpt-6-sol',
  messages: [
    { role: 'user', content: 'What is a deposition?' },
    { role: 'assistant', content: 'A deposition is sworn testimony taken outside of court...' },
    { role: 'user', content: 'How long do they typically last?' }
  ]
});

response = client.llm.v1.chat.create_completion(
    model='openai/gpt-6-sol',
    messages=[
        {'role': 'user', 'content': 'What is a deposition?'},
        {'role': 'assistant', 'content': 'A deposition is sworn testimony taken outside of court...'},
        {'role': 'user', 'content': 'How long do they typically last?'}
    ]
)

resp, _ := client.Llm.V1.Chat.NewCompletion(ctx, casedev.LlmV1ChatNewCompletionParams{
	Model: casedev.F("openai/gpt-6-sol"),
	Messages: casedev.F([]casedev.LlmV1ChatNewCompletionParamsMessage{
		{Role: casedev.F(casedev.LlmV1ChatNewCompletionParamsMessagesRoleUser), Content: casedev.F("What is a deposition?")},
		{Role: casedev.F(casedev.LlmV1ChatNewCompletionParamsMessagesRoleAssistant), Content: casedev.F("A deposition is sworn testimony taken outside of court...")},
		{Role: casedev.F(casedev.LlmV1ChatNewCompletionParamsMessagesRoleUser), Content: casedev.F("How long do they typically last?")},
	}),
})
fmt.Println(resp.Choices[0].Message.Content)

Streaming

Get responses token-by-token as they’re generated:

casedev llm:v1:chat create-completion \
  --model openai/gpt-6-sol \
  --message '{role: user, content: "Write a case summary."}' \
  --stream

const stream = await client.llm.v1.chat.createCompletion({
  model: 'openai/gpt-6-sol',
  messages: [{ role: 'user', content: 'Write a case summary.' }],
  stream: true
});

for await (const chunk of stream) {
  process.stdout.write(chunk.choices[0]?.delta?.content || '');
}

stream = client.llm.v1.chat.create_completion(
    model='openai/gpt-6-sol',
    messages=[{'role': 'user', 'content': 'Write a case summary.'}],
    stream=True
)

for chunk in stream:
    print(chunk.choices[0].delta.content or '', end='')

package main

import (
	"context"
	"net/http"

	casedev "github.com/CaseMark/casedev-go"
	"github.com/CaseMark/casedev-go/option"
)

func main() {
	client := casedev.NewClient()

	var httpResp *http.Response
	client.Llm.V1.Chat.NewCompletion(context.TODO(), casedev.LlmV1ChatNewCompletionParams{
		Model: casedev.F("openai/gpt-6-sol"),
		Messages: casedev.F([]casedev.LlmV1ChatNewCompletionParamsMessage{{
			Role:    casedev.F(casedev.LlmV1ChatNewCompletionParamsMessagesRoleUser),
			Content: casedev.F("Write a case summary."),
		}}),
		Stream: casedev.F(true),
	}, option.WithResponseInto(&httpResp))
	// Read httpResp.Body as SSE stream
}

Streaming HTTP failures

When an upstream provider rejects a streaming request before streaming begins, Case.dev returns a non-2xx JSON error instead of opening an SSE stream. The response preserves the upstream HTTP status and includes data.code: "UPSTREAM_PROVIDER_ERROR" with a generic message. Check this machine-readable code before interpreting the HTTP status. In particular, a 401 or 403 carrying UPSTREAM_PROVIDER_ERROR describes an upstream failure, not an invalid Case.dev API key. Do not automatically revoke or rotate the caller’s key based on that status alone. Provider response bodies are not returned. This describes pre-stream failures; failures after SSE has started follow the stream’s error handling.

Vision

Send images to models that support vision (GPT and Gemini):

Typescript

const response = await client.llm.v1.chat.createCompletion({
  model: 'openai/gpt-6-sol',
  messages: [
    {
      role: 'user',
      content: [
        { type: 'text', text: 'What medical equipment is visible in this image?' },
        { type: 'image_url', image_url: { url: 'https://example.com/exhibit-a.jpg' } }
      ]
    }
  ]
});

Usage and costs

Every response includes token counts and cost:

Response

{
  "usage": {
    "prompt_tokens": 1245,
    "completion_tokens": 387,
    "total_tokens": 1632,
    "cost": 0.004896
  }
}

Reduce costs: Use temperature: 0 for factual extraction. Try cheaper models like openai/gpt-6-luna or google/gemini-3.1-flash-lite for simpler tasks.

Common patterns

Deposition summary

casedev llm:v1:chat create-completion \
  --model openai/gpt-6-sol \
  --message '{role: system, content: "Summarize depositions with: 1. Key admissions 2. Timeline of events 3. Credibility issues 4. Contradictions with other testimony"}' \
  --message '{role: user, content: "<deposition text>"}' \
  --temperature 0.3 \
  --max-tokens 2000

const response = await client.llm.v1.chat.createCompletion({
  model: 'openai/gpt-6-sol',
  messages: [
    {
      role: 'system',
      content: `Summarize depositions with:
1. Key admissions
2. Timeline of events
3. Credibility issues
4. Contradictions with other testimony`
    },
    { role: 'user', content: depositionText }
  ],
  temperature: 0.3,
  max_tokens: 2000
});

response = client.llm.v1.chat.create_completion(
    model='openai/gpt-6-sol',
    messages=[
        {
            'role': 'system',
            'content': 'Summarize depositions with:\n1. Key admissions\n2. Timeline of events\n3. Credibility issues\n4. Contradictions with other testimony'
        },
        {'role': 'user', 'content': deposition_text}
    ],
    temperature=0.3,
    max_tokens=2000
)

resp, _ := client.Llm.V1.Chat.NewCompletion(ctx, casedev.LlmV1ChatNewCompletionParams{
	Model: casedev.F("openai/gpt-6-sol"),
	Messages: casedev.F([]casedev.LlmV1ChatNewCompletionParamsMessage{
		{
			Role:    casedev.F(casedev.LlmV1ChatNewCompletionParamsMessagesRoleSystem),
			Content: casedev.F("Summarize depositions with:\n1. Key admissions\n2. Timeline of events\n3. Credibility issues\n4. Contradictions with other testimony"),
		},
		{
			Role:    casedev.F(casedev.LlmV1ChatNewCompletionParamsMessagesRoleUser),
			Content: casedev.F(depositionText),
		},
	}),
	Temperature: casedev.F(0.3),
	MaxTokens:   casedev.F(int64(2000)),
})
fmt.Println(resp.Choices[0].Message.Content)

Contract clause extraction

casedev llm:v1:chat create-completion \
  --model openai/gpt-6-sol \
  --message '{role: system, content: "Extract all indemnification clauses. Return JSON: [{clause_text, page, party_protected}]"}' \
  --message '{role: user, content: "<contract text>"}' \
  --temperature 0

const response = await client.llm.v1.chat.createCompletion({
  model: 'openai/gpt-6-sol',
  messages: [
    {
      role: 'system',
      content: 'Extract all indemnification clauses. Return JSON: [{clause_text, page, party_protected}]'
    },
    { role: 'user', content: contractText }
  ],
  temperature: 0
});

response = client.llm.v1.chat.create_completion(
    model='openai/gpt-6-sol',
    messages=[
        {
            'role': 'system',
            'content': 'Extract all indemnification clauses. Return JSON: [{clause_text, page, party_protected}]'
        },
        {'role': 'user', 'content': contract_text}
    ],
    temperature=0
)

resp, _ := client.Llm.V1.Chat.NewCompletion(ctx, casedev.LlmV1ChatNewCompletionParams{
	Model: casedev.F("openai/gpt-6-sol"),
	Messages: casedev.F([]casedev.LlmV1ChatNewCompletionParamsMessage{
		{
			Role:    casedev.F(casedev.LlmV1ChatNewCompletionParamsMessagesRoleSystem),
			Content: casedev.F("Extract all indemnification clauses. Return JSON: [{clause_text, page, party_protected}]"),
		},
		{
			Role:    casedev.F(casedev.LlmV1ChatNewCompletionParamsMessagesRoleUser),
			Content: casedev.F(contractText),
		},
	}),
	Temperature: casedev.F(0.0),
})
fmt.Println(resp.Choices[0].Message.Content)

Medical record review

casedev llm:v1:chat create-completion \
  --model openai/gpt-6-astra \
  --message '{role: system, content: "You are a medical-legal expert. Identify standard-of-care deviations and timeline inconsistencies."}' \
  --message '{role: user, content: "<medical records>"}' \
  --max-tokens 5000

const response = await client.llm.v1.chat.createCompletion({
  model: 'openai/gpt-6-astra',
  messages: [
    {
      role: 'system',
      content: 'You are a medical-legal expert. Identify standard-of-care deviations and timeline inconsistencies.'
    },
    { role: 'user', content: medicalRecords }
  ],
  max_tokens: 5000
});

response = client.llm.v1.chat.create_completion(
    model='openai/gpt-6-astra',
    messages=[
        {
            'role': 'system',
            'content': 'You are a medical-legal expert. Identify standard-of-care deviations and timeline inconsistencies.'
        },
        {'role': 'user', 'content': medical_records}
    ],
    max_tokens=5000
)

resp, _ := client.Llm.V1.Chat.NewCompletion(ctx, casedev.LlmV1ChatNewCompletionParams{
	Model: casedev.F("openai/gpt-6-astra"),
	Messages: casedev.F([]casedev.LlmV1ChatNewCompletionParamsMessage{
		{
			Role:    casedev.F(casedev.LlmV1ChatNewCompletionParamsMessagesRoleSystem),
			Content: casedev.F("You are a medical-legal expert. Identify standard-of-care deviations and timeline inconsistencies."),
		},
		{
			Role:    casedev.F(casedev.LlmV1ChatNewCompletionParamsMessagesRoleUser),
			Content: casedev.F(medicalRecords),
		},
	}),
	MaxTokens: casedev.F(int64(5000)),
})
fmt.Println(resp.Choices[0].Message.Content)

Get Started

Platform

Resources

Parameters

Required

Optional

Messages

System prompts

Multi-turn conversations

Streaming

Streaming HTTP failures

Vision

Usage and costs

Common patterns

Deposition summary

Contract clause extraction

Medical record review

​Parameters

​Required

​Optional

​Messages

​System prompts

​Multi-turn conversations

​Streaming

​Streaming HTTP failures

​Vision

​Usage and costs

​Common patterns

​Deposition summary

​Contract clause extraction

​Medical record review

Parameters

Required

Optional

Messages

System prompts

Multi-turn conversations

Streaming

Streaming HTTP failures

Vision

Usage and costs

Common patterns

Deposition summary

Contract clause extraction

Medical record review