granite3-dense
ibm/granite3-dense:2b-instruct-q3_k_sThe IBM Granite 2B and 8B models are designed to support tool-based use cases and support for retrieval augmented generation (RAG), streamlining code generation, translation and bug fixing.
- Input → Output
- Text Text
- Access
- Token Exchange
- Context window
- 4,096 tokens
- Model tags
- 33 versions available
From model to application
Quick Start
Get your API key
Create a key in the ShareAI Console, then save it as an environment variable on your server.
Create API keyexport SHAREAI_API_KEY="your-api-key"Choose your endpoint
Send a POST request with your API key in the Authorization header.
POSThttps://api.shareai.now/api/v1/chat/completionsThis tag’s endpoint has not been confirmed. Check the API reference and your Console before sending a request.
Make your first request
The examples use your selected tag and show its request format. Run them on your server to keep your API key private. Model availability and account access are checked when you send a request.
curl --fail-with-body --request POST 'https://api.shareai.now/api/v1/chat/completions' \ --header "Authorization: Bearer ${SHAREAI_API_KEY}" \ --header 'Content-Type: application/json' \ --data-raw '{ "model": "ibm/granite3-dense:2b-instruct-q3_k_s", "messages": [ { "role": "user", "content": "Hello!" } ], "stream": false }'import json import os import urllib.request payload = { "model": "ibm/granite3-dense:2b-instruct-q3_k_s", "messages": [ { "role": "user", "content": "Hello!", }, ], "stream": False, } request = urllib.request.Request("https://api.shareai.now/api/v1/chat/completions", data=json.dumps(payload).encode(), headers={ "Authorization": f"Bearer {os.environ['SHAREAI_API_KEY']}", "Content-Type": "application/json", }, method="POST", ) with urllib.request.urlopen(request, timeout=180) as response: print(json.dumps(json.load(response), indent=2))package main import ( "bytes" "encoding/json" "fmt" "io" "net/http" "os" "time" ) func main() { if err := run(); err != nil { fmt.Fprintln(os.Stderr, err) os.Exit(1) } } func run() error { key := os.Getenv("SHAREAI_API_KEY") if key == "" { return fmt.Errorf("set SHAREAI_API_KEY") } body := []byte(`{ "model": "ibm/granite3-dense:2b-instruct-q3_k_s", "messages": [ { "role": "user", "content": "Hello!" } ], "stream": false }`) request, err := http.NewRequest(http.MethodPost, "https://api.shareai.now/api/v1/chat/completions", bytes.NewReader(body)) if err != nil { return err } request.Header.Set("Authorization", "Bearer " + key) request.Header.Set("Content-Type", "application/json") client := &http.Client{Timeout: 180 * time.Second} response, err := client.Do(request) if err != nil { return err } defer response.Body.Close() if response.StatusCode < 200 || response.StatusCode >= 300 { detail, _ := io.ReadAll(io.LimitReader(response.Body, 1<<20)) return fmt.Errorf("ShareAI HTTP %d: %s", response.StatusCode, detail) } var result any if err := json.NewDecoder(response.Body).Decode(&result); err != nil { return err } encoder := json.NewEncoder(os.Stdout) encoder.SetIndent("", " ") return encoder.Encode(result) }Save as main.go and run go run main.go. This example uses the Go standard library.
Set up Rust dependencies
In your Cargo project, add these dependencies. Save the example as src/main.rs, then run cargo run.
cargo add [email protected] --no-default-features --features blocking,json,rustls-tls cargo add serde_json@1fn main() -> Result<(), Box<dyn std::error::Error>> { let api_key = std::env::var("SHAREAI_API_KEY")?; let payload: serde_json::Value = serde_json::from_str(r#"{ "model": "ibm/granite3-dense:2b-instruct-q3_k_s", "messages": [ { "role": "user", "content": "Hello!" } ], "stream": false }"#)?; let client = reqwest::blocking::Client::builder() .timeout(std::time::Duration::from_secs(180)) .build()?; let response = client.post(r#"https://api.shareai.now/api/v1/chat/completions"#) .bearer_auth(api_key) .json(&payload) .send()? .error_for_status()?; let result: serde_json::Value = response.json()?; println!("{}", serde_json::to_string_pretty(&result)?); Ok(()) }const payload = { "model": "ibm/granite3-dense:2b-instruct-q3_k_s", "messages": [ { "role": "user", "content": "Hello!" } ], "stream": false }; const response = await fetch("https://api.shareai.now/api/v1/chat/completions", { method: 'POST', headers: { Authorization: `Bearer ${process.env.SHAREAI_API_KEY}`, 'Content-Type': 'application/json', }, body: JSON.stringify(payload), }); if (!response.ok) throw new Error(await response.text()); console.log(JSON.stringify(await response.json(), null, 2));
Token Exchange
Available through Token Exchange
Use the network with your token balance. Your account’s consumption settings determine how requests are billed.
A little more detail
Frequently asked questions
What is granite3-dense?
The IBM Granite 2B and 8B models are designed to support tool-based use cases and support for retrieval augmented generation (RAG), streamlining code generation, translation and bug fixing.
How much does granite3-dense cost?
This tag is available through Token Exchange. Your token balance and account consumption settings determine access and billing. Check your Console for your balance and eligibility.
What inputs and outputs does granite3-dense support?
Inputs: Text. Outputs: Text. These capabilities apply to the selected model tag. Check the endpoint documentation for supported request formats.
What is the context length of granite3-dense?
This tag has a context window of 4,096 tokens. Context size and pricing thresholds are separate. Request limits can be lower for a particular deployment.
Does granite3-dense support tool calling and structured outputs?
Tool calling: Supported.
How do I use granite3-dense with ShareAI?
Create an API key in the ShareAI Console, choose a model tag, and send an authenticated request to the endpoint in Quick Start. Your key must have access to the chosen model. Keep your key on your server.