MiniLLMLib-RS

A minimalist, async-first Rust library for LLM interactions with streaming support.

Features

• Async-first: Built on Tokio for high-performance async operations
• Streaming Support: First-class SSE streaming for real-time responses
• Conversation Trees: ChatNode provides tree-based conversation structure with branching
• Tree Manipulation: detach(), merge(), tree iterators (depth-first, breadth-first, leaves)
• Template Substitution: Format kwargs with {placeholders} in messages
• Thread Serialization: Save/load conversation threads to/from JSON files
• Cost Tracking: OpenRouter usage accounting with callbacks
• Multimodal: Support for images and audio in messages
• JSON Repair: Robust handling of malformed JSON from LLM outputs
• OpenRouter Compatible: Works with OpenRouter, OpenAI, and any OpenAI-compatible API
• Retry with Backoff: Built-in exponential backoff and retry logic
• Provider Routing: OpenRouter provider settings (sort, ignore, data collection)

Installation

Add to your Cargo.toml:

[dependencies]
minillmlib = "0.2"
tokio = { version = "1", features = ["rt-multi-thread", "macros"] }

Quick Start

use minillmlib::{ChatNode, GeneratorInfo};

#[tokio::main]
async fn main() -> minillmlib::Result<()> {
    // Load .env and configure logging
    minillmlib::init();

    // Create a generator for OpenRouter
    let generator = GeneratorInfo::openrouter("google/gemini-2.5-flash-lite");

    // Start a conversation
    let root = ChatNode::root("You are a helpful assistant.");
    let response = root.chat("Hello!", &generator).await?;

    println!("Assistant: {}", response.text().unwrap_or_default());
    Ok(())
}

Environment Variables

Set your API key in a .env file or environment:

OPENROUTER_API_KEY=sk-or-v1-your-key-here
# Or for direct OpenAI:
OPENAI_API_KEY=sk-your-key-here

Usage Examples

Basic Completion

use minillmlib::{ChatNode, GeneratorInfo, CompletionParameters, NodeCompletionParameters};

let generator = GeneratorInfo::openrouter("anthropic/claude-3.5-sonnet");
let root = ChatNode::root("You are helpful.");
let user = root.add_user("What is 2+2?");

// With custom parameters
let params = NodeCompletionParameters::new()
    .with_params(
        CompletionParameters::new()
            .with_temperature(0.0)
            .with_max_tokens(100)
    );

let response = user.complete(&generator, Some(&params)).await?;
println!("{}", response.text().unwrap());

Streaming

let root = ChatNode::root("You are helpful.");
let user = root.add_user("Tell me a story.");

let mut stream = user.complete_streaming(&generator, None).await?;

while let Some(chunk) = stream.next_chunk().await {
    print!("{}", chunk?.delta);
}

Multi-turn Conversation

let root = ChatNode::root("You are helpful.");

// First turn
let response1 = root.chat("My name is Alice.", &generator).await?;

// Second turn - context is preserved
let response2 = response1.chat("What's my name?", &generator).await?;
// Response will mention "Alice"

Image Input

use minillmlib::{ChatNode, GeneratorInfo, ImageData, MessageContent, Message, Role};

let generator = GeneratorInfo::openrouter("google/gemini-2.5-flash-lite");
let image = ImageData::from_file("./image.jpg")?;

let content = MessageContent::with_images("Describe this image.", &[image]);
let root = ChatNode::root("You are helpful.");
let user = root.add_child(ChatNode::new(Message {
    role: Role::User,
    content,
    name: None,
    tool_call_id: None,
    tool_calls: None,
}));

let response = user.complete(&generator, None).await?;

Audio Input

use minillmlib::{AudioData, MessageContent};

let audio = AudioData::from_file("./audio.mp3")?;
let content = MessageContent::with_audio("Transcribe this audio.", &[audio]);

JSON Response with Repair

let params = NodeCompletionParameters::new()
    .with_parse_json(true)           // Enable JSON repair
    .with_crash_on_refusal(true)     // Retry if no valid JSON
    .with_retry(3);                  // Number of retries

let response = user.complete(&generator, Some(&params)).await?;
// response.text() will contain valid, repaired JSON

Retry with Exponential Backoff

let params = NodeCompletionParameters::new()
    .with_retry(5)
    .with_exp_back_off(true)
    .with_back_off_time(1.0)    // Start with 1 second
    .with_max_back_off(30.0)    // Max 30 seconds
    .with_crash_on_empty(true); // Retry on empty responses

Force Prepend (Constrained Generation)

// Force the model to start its response with specific text
let params = NodeCompletionParameters::new()
    .with_force_prepend("Score: ");

// Response will start with "Score: " followed by the model's completion

OpenRouter Provider Settings

OpenRouter routing is provider-specific, so it's attached via with_openrouter_routing (which carries it under the request's provider key); non-OpenRouter providers simply ignore it.

use minillmlib::{CompletionParameters, ProviderSettings};

let routing = ProviderSettings::new()
    .sort_by_throughput()                              // or .sort_by_price()
    .deny_data_collection()
    .with_ignore(vec!["SambaNova".to_string()]);       // Exclude providers

let params = CompletionParameters::new()
    .with_openrouter_routing(routing);

Prompt Caching (provider-agnostic)

Mark what to cache on the tree; the provider decides the wire (Anthropic emits cache_control, OpenAI auto-caches). Switch the provider and the same code works.

let root = ChatNode::root(big_system_prompt);
root.cache_breakpoint();                 // cache just the system prompt
// ...or NodeCompletionParameters::new().with_cache(true) to cache the whole prefix

// Warm the cache before an agent run (cheap to call repeatedly):
let warm_cost = some_node.ensure_cached(&generator, None).await?;

// Clear marks:
root.clear_cache_breakpoint();           // one node
root.clear_all_cache_breakpoints();      // whole tree

Cache tokens are priced with distinct read/write rates (cache reads are ~0.1× input; cache writes a ~1.25× premium):

let price = TokenPrice::new(1.0, 5.0)        // $/Mtok input, output
    .with_cache_rates(0.1, 1.25);            // $/Mtok cache-read, cache-write

Custom/Extra Parameters

// Pass arbitrary parameters to the API
let params = CompletionParameters::new()
    .with_extra("custom_param", serde_json::json!(42))
    .with_extra("another", serde_json::json!({"nested": "value"}));

Pretty Print Conversations

use minillmlib::{pretty_messages, format_conversation, PrettyPrintConfig};

let root = ChatNode::root("You are helpful.");
let user = root.add_user("Hello");
let assistant = user.add_assistant("Hi there!");

// Default formatting
let pretty = format_conversation(&assistant);
// Output: "SYSTEM: You are helpful.\n\nUSER: Hello\n\nASSISTANT: Hi there!"

// Custom formatting
let config = PrettyPrintConfig::new("[SYS] ", "\n[USR] ", "\n[AST] ");
let pretty = pretty_messages(&assistant, Some(&config));

Template Substitution (Format Kwargs)

use minillmlib::ChatNode;

// Create a reusable prompt template
let root = ChatNode::root("You are {bot_name}, a {style} assistant.");
root.set_format_kwarg("bot_name", "Claude");
root.set_format_kwarg("style", "helpful");

let user = root.add_user("Hi {bot_name}!");

// Get formatted messages with placeholders replaced
let formatted = user.formatted_thread();
// Messages now contain "You are Claude, a helpful assistant." etc.

Save and Load Conversation Threads

use minillmlib::ChatNode;

// Build a conversation
let root = ChatNode::root("You are helpful.");
root.set_format_kwarg("name", "Alice");
let user = root.add_user("Hello {name}!");
let assistant = user.add_assistant("Hi there!");

// Save to JSON file
assistant.save_thread("conversation.json")?;

// Load from JSON file (returns root and leaf)
let (loaded_root, loaded_leaf) = ChatNode::from_thread_file("conversation.json")?;

// Or load from JSON string
let json = r#"{"prompts": [{"role": "system", "content": "Hello"}], "required_kwargs": {}}"#;
let (root, leaf) = ChatNode::from_thread_json(json)?;

Tree Manipulation

use minillmlib::ChatNode;

// Navigate to root from any node
let root = some_deep_node.get_root();

// Detach a subtree
let subtree = node.detach();  // node is now a new root

// Merge trees
let merged = tree1_leaf.merge(&tree2_leaf);  // tree2's root becomes child of tree1_leaf

// Iterate over tree
for node in root.iter_depth_first() {
    println!("{}", node.text().unwrap_or_default());
}

// Get all leaves
let leaves = root.iter_leaves();

// Count nodes
let count = root.node_count();

Cost Tracking (OpenRouter)

use minillmlib::{ChatNode, GeneratorInfo, NodeCompletionParameters, CostInfo};
use std::sync::{Arc, Mutex};

let generator = GeneratorInfo::openrouter("google/gemini-2.5-flash-lite");

// Track costs across multiple requests
let total_cost = Arc::new(Mutex::new(0.0));
let cost_tracker = total_cost.clone();

let params = NodeCompletionParameters::new()
    .with_cost_tracking(true)
    .with_cost_callback(move |info: CostInfo| {
        *cost_tracker.lock().unwrap() += info.cost;
        println!("Request cost: {} credits", info.cost);
        println!("Tokens: {} prompt, {} completion", 
            info.prompt_tokens, info.completion_tokens);
    });

let root = ChatNode::root("You are helpful.");
let user = root.add_user("Hello!");
let response = user.complete(&generator, Some(&params)).await?;

println!("Total spent: {} credits", *total_cost.lock().unwrap());

API Reference

Core Types

GeneratorInfo Methods

// Pre-configured providers
GeneratorInfo::openrouter(model)         // OpenRouter (OpenAI wire, native USD cost)
GeneratorInfo::openai(model)             // OpenAI (token-only; price via with_token_price)
GeneratorInfo::anthropic(model)          // Native Anthropic /v1/messages, x-api-key auth
GeneratorInfo::claude_subscription(model)// Anthropic wire, Claude Pro/Max OAuth token
GeneratorInfo::custom(name, url, model)  // Custom OpenAI-compatible endpoint

// Auth builder methods
.with_api_key(key)                       // provider chooses header (Bearer / x-api-key)
.with_api_key_from_env("ENV_VAR")
.with_bearer_token(token)                // OAuth / subscription bearer token
.with_bearer_token_from_env("ENV_VAR")

// Other builder methods
.with_token_price(TokenPrice::new(in_per_mtok, out_per_mtok)) // cost estimate for token-only providers
.with_provider(Arc::new(MyProvider))     // swap the wire dialect
.with_header(name, value)
.with_vision()
.with_audio()
.with_max_context(length)
.with_default_params(params)

Claude Subscription (use your Pro/Max plan)

A Claude Pro/Max subscription OAuth token authenticates against the native Anthropic API the same way an API key does, but draws on your subscription's rolling quota (the 5-hour / 7-day window) instead of pay-as-you-go API billing.

claude_subscription resolves the token in this order:

1. the ANTHROPIC_AUTH_TOKEN env var, if set (explicit override, e.g. from claude setup-token; you keep it fresh);
2. otherwise the live Claude Code credential at ~/.claude/.credentials.json (claudeAiOauth.accessToken), which Claude Code keeps refreshed, so if you're logged into Claude Code with your subscription, it just works.

use minillmlib::{ChatNode, GeneratorInfo, TokenPrice};

// Anthropic returns token counts but no dollar cost, so set a price for a
// resolved cost ESTIMATE (otherwise tracking reports `Unpriced`).
let generator = GeneratorInfo::claude_subscription("claude-haiku-4-5")
    .with_token_price(TokenPrice::new(1.0, 5.0)); // $/Mtok in, $/Mtok out

let root = ChatNode::root("You are helpful.");
let response = root.chat("Hello!", &generator).await?;

Subscription vs Console. A subscription token (from Claude Code) bills your Pro/Max plan. A Console/API OAuth token (e.g. from the ant CLI) bills your API account, not the subscription; for Console use an API key via GeneratorInfo::anthropic(model). Verify which bucket you're hitting by the response's rate-limit headers: subscription returns anthropic-ratelimit-unified-5h-*; the API tier returns anthropic-ratelimit-input-tokens-limit.

CompletionParameters

NodeCompletionParameters

ProviderSettings (OpenRouter)

CLI Tool

The library includes a CLI for JSON repair:

# Repair JSON from file
minillmlib-cli input.json

# Repair JSON from stdin
echo '{"key": "value",}' | minillmlib-cli

Running Tests

# Default: all offline tests (unit + offline integration). No API calls, free.
cargo test

# Unit tests only (fast)
cargo test --lib

# Live integration tests (REAL, billed API calls): opt in with the `live` feature.
# Reads OPENROUTER_API_KEY, ANTHROPIC_API_KEY, ANTHROPIC_AUTH_TOKEN from the env
# (or a .env); each live test skips gracefully if its key is absent.
cargo test --features live

# Run with output
cargo test -- --nocapture

Without --features live, every network test skips, so cargo test is free, offline, and deterministic even when real keys are present in your environment.

License

MIT License - see LICENSE for details.

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.