Lynkr

An LLM Gateway which optimises your token usage.

84% fewer tokens on JSON tool results. 53% fewer tokens on tool-heavy requests. Sub-300ms semantic cache hits. Zero code changes.

Numbers from the bundled benchmark against LiteLLM on identical free local backends — run it yourself: node benchmark-tier-routing.js. It doubles as a 19-scenario routing regression harness (currently 12/12 correctness checks), and MODE=routing runs a routing-only head-to-head that judges both proxies on the same acceptable-tier sets — including LiteLLM's Auto Router v2. How it works →

Agentic benchmark (Terminal-Bench core, 80 tasks, Oct 2026): Lynkr 42/80 for $1.54 vs openrouter/auto 44/80 for $4.72 on the same VM and evening — tied on accuracy within single-run noise (Lynkr's three runs: 39, 39, 42) at ~30% of the cost; vLLM Semantic Router on the same models/hosts: 25/80 (34 agent timeouts from unbounded upstream stalls and per-turn CPU classification). Full comparison, fairness notes and what it changed in Lynkr →

Third-party benchmark: on RouterArena (ICLR 2026, 8,400 queries) Lynkr's routing scores 67.65 arena / 68.41% accuracy at $0.29 per 1K queries with 92.38 robustness — above GPT-5's built-in router and NotDiamond at a fraction of their cost. Methodology & caveats →

---

🚀 New: Wrap Mode for AI Coding Tools

Use Lynkr's routing with your AI coding assistant — maximize your subscription value:

npm install -g lynkr


# Claude Code Pro/Max
lynkr wrap claude

Wrapping gives you:

• ✅ Tier routing (send simple tasks to free Ollama, complex to your subscription/API)
• ✅ Sticky sessions: one routing decision per conversation via content fingerprinting, with automatic escalation when the task outgrows the model
• ✅ TOON/RTK compression (84% token reduction on large JSON tool outputs)
• ✅ Semantic caching (sub-300ms cache hits, 0 tokens billed)
• ✅ Your subscription is used only where it counts — simple requests route to your own local/API models, so subscription capacity goes to the work that needs it
• ✅ Works with OAuth (Claude, Copilot, Cursor) or API keys (Aider, Codex)

Full wrap guide →

---

Quick Start (2 Minutes)

1. Install Lynkr

npm install -g lynkr

2. Configure Lynkr

The fastest path is the interactive wizard:

lynkr init

It asks four questions — usage mode (Claude Pro/Max via wrap, or direct API keys), tier picks for SIMPLE/MEDIUM/COMPLEX/REASONING across the 14 supported providers, credentials for what you chose, and a few routing-intelligence knobs — then writes a fully-populated .env with sensible production defaults for everything else (caching, compression, policy budgets, MCP sandbox, rate limiting).

Useful flags:

lynkr init --force                # overwrite an existing .env

See docs/init.md for the full wizard reference.

If you'd rather configure by hand, the manual options below still work — copy .env.example to .env and edit it directly:

Option A: Free & Local (Ollama) - Recommended for Testing

# Install Ollama first: https://ollama.com
ollama pull qwen2.5-coder:latest

Then start Lynkr:

lynkr start

3. Connect Your Tool

Cursor IDE

• Settings → Models → Override Base URL
• Set to: http://localhost:8081/v1
• API Key: any-value

Codex CLI

Edit ~/.codex/config.toml:

model_provider = "lynkr"

[model_providers.lynkr]
base_url = "http://localhost:8081/v1"
wire_api = "responses"

✅ Done! Your AI tool now uses your chosen provider.

---

Common Startup Errors

Error: unable to determine transport target for "pino-pretty"

Problem: You're running an older version (< 9.3.0).

Solution: Update to the latest version:

npm install -g lynkr@latest

If you must use an older version, set NODE_ENV=production before starting.

Warning: Missing tier configuration: TIER_SIMPLE, TIER_MEDIUM...

This is just a warning - you can ignore it. Tier routing is optional.

To remove the warning, add to .env:

TIER_SIMPLE=ollama:qwen2.5-coder:latest
TIER_MEDIUM=ollama:qwen2.5-coder:latest
TIER_COMPLEX=ollama:qwen2.5-coder:latest
TIER_REASONING=ollama:qwen2.5-coder:latest

Warning: FALLBACK_PROVIDER='databricks' is enabled but missing credentials

Solution: Add to .env:

FALLBACK_ENABLED=false

Error: connect ECONNREFUSED ::1:11434 (Ollama)

Problem: Ollama is not running.

Solution:

ollama serve

Keep this terminal open, and start Lynkr in a new terminal.

Error: Connection refused or 404 Not Found

Problem: Lynkr is not running or wrong port.

Solution: Check Lynkr is running on the correct port:

curl http://localhost:8081/

Should return: {"service":"Lynkr","version":"9.x.x","status":"running"}

---

Why Lynkr?

AI coding tools lock you into one provider and send every token raw. Lynkr breaks both locks.

Claude Code / Cursor / Codex / Cline / Continue
                    ↓
                  Lynkr
          ┌─────────────────────┐
          │  Strip unused tools  │  ← 53% fewer tokens on tool calls
          │  Compress JSON blobs │  ← 84% on large tool results
          │  Semantic cache      │  ← <300ms hits, 0 tokens billed
          │  Route by complexity │  ← cheap model for simple, cloud for hard
          │  Learn from outcomes │  ← kNN + bandit + auto-calibration
          └─────────────────────┘
                    ↓
    Ollama | Bedrock | Azure | Moonshot | OpenRouter | OpenAI

What you get:

• ✅ 53% fewer tokens on tool-heavy requests (Claude Code, Cursor sessions)
• ✅ 84% compression on large JSON tool results (grep, file reads, test output)
• ✅ Semantic cache serves repeated queries in under 300ms with 0 tokens billed
• ✅ Automatic tier routing — simple questions go to cheap models, complex ones escalate; sessions stick to one model until the task genuinely outgrows it
• ✅ A closed learning loop — every outcome trains a kNN router and bandit, and tier thresholds re-calibrate nightly from your own traffic
• ✅ Route through your company's infrastructure (Databricks, Azure, Bedrock)
• ✅ Zero code changes — just change one environment variable

---

Supported Providers

Provider	Type	Example Models	Cost
Ollama	Local	qwen2.5-coder, deepseek-coder, llama3	Free
llama.cpp	Local	Any GGUF model	Free
LM Studio	Local	Local models with GUI	Free
OpenRouter	Cloud	GPT-4o, Claude 3.5, Llama 3, Gemini	$
OrcaRouter	Cloud	Chat, reasoning, image, video (one endpoint, live catalog)	$ (zero-markup)
AWS Bedrock	Cloud	Claude, Llama, Mistral, Titan	$$
Databricks	Cloud	Claude Sonnet 4.5, Opus 4.6	$$$
Azure OpenAI	Cloud	GPT-4o, o1, o3	$$$
Azure Anthropic	Cloud	Claude Sonnet, Opus	$$$
OpenAI	Cloud	GPT-4o, o3-mini	$$$
Atlas Cloud	Cloud	Qwen, DeepSeek, and other OpenAI-compatible models	$-$$$
DeepSeek	Cloud	DeepSeek R1, Reasoner	$
Z.ai	Cloud	GLM-4.7, GLM-4.5-Air	$
Moonshot AI	Cloud	Kimi K2.6, Kimi K3	$
Baidu Qianfan	Cloud	ERNIE 4.5 Turbo, ERNIE X1.1	$ (unverified — not yet probed against a live key)
Fireworks AI	Cloud	Kimi K2, GLM-5, DeepSeek V3 (serverless)	$ (unverified — not yet probed against a live key)

4 local providers for 100% offline, free usage. 14+ cloud providers for scale.

---

Advanced: Tier Routing (Save Even More)

Route different request types to different models automatically:

# .env file
MODEL_PROVIDER=ollama
FALLBACK_ENABLED=false

# Use small/fast models for simple tasks
TIER_SIMPLE=ollama:qwen2.5:3b

# Use medium models for normal coding
TIER_MEDIUM=ollama:qwen2.5:7b

# Use powerful models for complex architecture
TIER_COMPLEX=ollama:deepseek-r1:14b
TIER_REASONING=ollama:deepseek-r1:14b

# Optional: Limits (remove for unlimited) for long conversations
POLICY_MAX_STEPS=50
POLICY_MAX_TOOL_CALLS=100

Lynkr analyzes each request and routes it to the appropriate tier. Simple questions use fast models. Complex refactoring uses powerful models. The scorer combines anchor-embedding classification (WS7, payload-invariant) with an LLM difficulty classifier (Phase 6, added 2026-07-19) that catches topic-vs-difficulty confounding — list the exports from this file correctly routes to MEDIUM instead of getting boosted to REASONING by "technical vocabulary" alone.

Result: 70-90% of requests use cheaper/faster models. Only hard problems hit expensive models. 15× reduction in expensive-tier over-routing vs the anchor-only baseline on the 381-prompt eval set (0.6% vs ~15%).

Tier configuration is strictly authoritative — bandit exploration is constrained to the models you've listed in TIER_*, and multi-turn conversations score with a recency-weighted sliding window so context isn't lost on short follow-ups. Conversations get a content-fingerprint session id (clients like Claude Code send none), the decision pins for the session, and a guarded escape ladder (risk keywords, force phrases, score drift, context overflow) re-escalates the moment a task outgrows its model. Full pipeline: docs/routing-intelligence.md · intent scorer: docs/intent-window-routing.md · verify any change: docs/benchmarking.md.

---

Common Issues & Fixes

Issue	Solution
"Service temporarily overloaded"	Ollama model too large for RAM. Use smaller model or increase --max-old-space-size
"Route not found: HEAD /"	Ignore - harmless health check from Claude Code
"Hallucinated tool calls"	Normal - Lynkr automatically filters invalid tools
"Safe Command DSL blocked"	Add POLICY_SAFE_COMMANDS_ENABLED=false to .env
"spawn graphify ENOENT"	Optional feature. Set CODE_GRAPH_ENABLED=false in .env (see Advanced Features section for installation)
Slow first request (20+ sec)	Ollama loading model into memory. Add OLLAMA_KEEP_ALIVE=30m in Ollama config
No response after N turns	Remove POLICY_MAX_STEPS and POLICY_MAX_TOOL_CALLS from .env (unlimited by default in v9.3.0+)

---

Advanced Features

Real-time SSE streaming (all tiers)

Responses stream token-by-token instead of arriving all at once — including through the tier router. Two mechanisms, both on by default:

• Native passthrough: when the upstream already speaks Anthropic SSE (Anthropic endpoints, Z.AI, Ollama v0.14+), Lynkr pipes the bytes straight through with backpressure. Kill switch: LYNKR_NATIVE_PASSTHROUGH=false.
• Cross-format transform: OpenAI-format upstreams (OpenAI, Azure OpenAI, OpenRouter, Databricks, llama.cpp, LM Studio) are reshaped into Anthropic events in flight — including reassembling tool-call argument fragments into clean tool_use blocks. Kill switch: LYNKR_STREAM_TRANSFORM=false.

The LYNKR_VISIBLE_ROUTING badge streams too (injected as the first content block), and telemetry is recorded on stream close with real token counts. Fallback safety: if an upstream fails before the first byte, the request falls back to the buffered path automatically.

# Ollama thinking models (MiniMax): streaming skips the <think>-leak repair.
# If you see raw <think> text in responses, buffer that provider instead:
LYNKR_OLLAMA_BUFFER_RESPONSES=true   # default true; false = stream Ollama

Token Optimization (60-80% savings)

# Enable all optimizations
PROMPT_CACHE_ENABLED=true
SEMANTIC_CACHE_ENABLED=true

Always-on (no config): RTK tool-result compression (test/git/grep/lint/build/JSON output), MCP tool dedup (drops built-in WebSearch/WebFetch when an Exa/Tavily MCP tool is present), and request bypass (Claude CLI Warmup / title-extraction calls are answered locally, never hitting a provider).

Optional terse-output mode to cut output tokens:

CAVEMAN_ENABLED=true        # off by default — nudges the model to be concise
CAVEMAN_LEVEL=lite          # lite | full | ultra

Built-in dashboard

Open http://localhost:8081/dashboard while Lynkr is running. No setup, reads the local telemetry store:

• Spend & savings — actual cost vs what the same traffic would have cost on the flagship model (the counterfactual no pass-through gateway can show, since they neither pick your model nor see local-model traffic)
• Tier mix — daily request breakdown across SIMPLE/MEDIUM/COMPLEX/REASONING
• Routing accuracy — over-/under-provisioned request counts, a self-audit of the tier router's decisions
• Request logs — filterable by provider, tier, and errors, with latency, tokens, and cost per request; every session id links to a drill-down
• Provider health — configured providers, credential warnings, circuit breaker states
• Insights — analyzer findings ranked by past overspend (auditable against spend already recorded, never projected savings): dead tool schemas paying rent, harness side-traffic served on expensive tiers, weak prompt-cache hit ratios, and Wilson-bounded tier-downsize verdicts. Each finding ships its evidence rows, the fix, the caveat, and how the figure was computed. "Not measured yet" is rendered distinctly from zero — an un-run scan is not an all-clear.
• Explore — a Metric × Dimension (× Stack) pivot over routing telemetry (spend/tokens/requests/sessions by provider, model, tier, request type, day…) with preset views, period-over-period deltas, and CSV export
• Session drill-down — per-model cost mix and an input-context-per-call chart where distillation and cache behavior are directly visible
• Cache receipts — projected switch/hold economics from cache-aware routing next to the measured companion (dollars saved by cache reads actually served)

JSON APIs behind it (/dashboard/api/overview|usage|routing|logs|recommendations|analytics|sessions/:id|statusline) if you want the raw numbers.

Status line for Claude Code

One line after every turn — routed tier, served model, today's spend, and cache re-read share — with zero token cost (out-of-band, never enters model context). Add to ~/.claude/settings.json:

"statusLine": { "type": "command", "command": "lynkr-statusline" }

◆ Opus 4.8 · MEDIUM → minimax-m3:cloud (ollama) · today $0.06 · cache 67%

Always prints exactly one line and always exits 0 — a down proxy can never break the harness. Set LYNKR_PORT if Lynkr isn't on 8081.

Cost tracking & model pricing

Per-request cost is computed from a model-pricing registry (LiteLLM → models.dev, cached 24h) and recorded in telemetry. Models the registry doesn't know record cost_usd=null (logged once) rather than a fabricated price. Pin prices for unknown models:

# Per-1M-token USD prices, JSON keyed by model name
MODEL_PRICE_OVERRIDES={"my-model":{"input":0.5,"output":1.5}}

Memory System (Titans-inspired)

MEMORY_ENABLED=true
MEMORY_TTL=3600000  # 1 hour

Load Shedding & Resilience

LOAD_SHEDDING_ENABLED=true
LOAD_SHEDDING_HEAP_THRESHOLD=0.85

Admin Hot-Reload (no restart needed)

curl -X POST http://localhost:8081/v1/admin/reload

Code Intelligence (Optional - Graphify)

Graphify provides AST-based code analysis for smarter routing decisions.

Installation (Rust required):

# Install Rust if not already installed
curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh
source $HOME/.cargo/env

# Build and install graphify
git clone https://github.com/safishamsi/graphify
cd graphify
cargo build --release
sudo cp target/release/graphify /usr/local/bin/

# Verify installation
graphify --version

Enable in .env:

CODE_GRAPH_ENABLED=true
CODE_GRAPH_WORKSPACE=/path/to/your/project  # Optional, defaults to cwd

Features:

• AST-based complexity scoring
• Structural code analysis (19 languages supported)
• Enhanced routing decisions based on code structure

Note: Graphify is completely optional. If not installed, Lynkr falls back to simpler complexity analysis.

---

Installation Methods

NPM (recommended)

npm install -g lynkr

One-line installer

curl -fsSL https://raw.githubusercontent.com/Fast-Editor/Lynkr/main/install.sh | bash

Homebrew (macOS / Linux)

brew tap fast-editor/lynkr
brew install lynkr
lynkr --version

Upgrade later with brew update && brew upgrade lynkr. The formula tracks the latest lynkr npm release automatically.

Docker

git clone https://github.com/Fast-Editor/Lynkr.git
cd Lynkr
docker-compose up -d

From source

git clone https://github.com/Fast-Editor/Lynkr.git
cd Lynkr
npm install
cp .env.example .env
npm start

---

Documentation

Guide	Description
Installation	All installation methods
Provider Setup	Configuration for all 12+ providers
Claude Code	Claude Code CLI integration
Cursor IDE	Cursor setup + troubleshooting
Codex CLI	Codex configuration
Tier Routing	Smart model routing by complexity
Token Optimization	60-80% cost reduction
Troubleshooting	Common issues and solutions
API Reference	REST API endpoints
Production	Enterprise deployment

---

Benchmark Results

Head-to-head against LiteLLM on the same backends (Ollama minimax-m2.5, Moonshot, Azure OpenAI), 9 scenarios across 4 feature categories. Apples-to-apples comparison is Lynkr vs LiteLLM billed tokens on the same scenario. Run with node benchmark-tier-routing.js.

Runs: token benchmarks — June 5, 2026 (Lynkr v9.3.2 · LiteLLM v1.87.1). Tier routing head-to-head — re-run July 19, 2026 (Lynkr working tree w/ Phase-6 classifier + config B · LiteLLM v1.94.0.dev1 Auto Router v2). macOS, Apple Silicon.

Token reduction (vs LiteLLM, same model & prompt)

Mechanism	Lynkr	LiteLLM	Result
TOON compression (60-item grep JSON)	427 tokens · $0.009	3,458 tokens · $0.018	87.6% fewer tokens, 50% cheaper

Lynkr binary-compresses large JSON tool results (TOON) in-process, with no added latency.

Semantic cache

	Tokens billed	Response time
First call (cold)	2,857	1,891ms
Second call — paraphrased, cache hit	0 (served from cache)	171ms (11× faster)

Near-identical prompts return cached responses in 171ms. Zero model tokens billed on a cache hit.

Tier routing — vs LiteLLM Auto Router v2 (re-run July 19, 2026)

LiteLLM v1.94 shipped a native complexity router (auto_router/complexity_router) with the same four tier names Lynkr uses. Head-to-head on the same backends with identical prompts, both proxies live, both judged against the same acceptable-tier sets (11 routing scenarios, MODE=routing node benchmark-tier-routing.js, LiteLLM config in litellm-autorouter-v2.yaml):

Router	Routing-correct	Notes
Lynkr (Phase-6 classifier + config B)	11/11 ✅	anchor embedding + local LLM classifier reconcile + FORCE_REASONING patterns + risk-remap; ~500ms warm classifier call, cached
LiteLLM v2 — heuristic (default)	4/11	every miss under-routed — banking security analysis, whole-pipeline refactor, prod auth-file fix, autonomous agentic loop all sent to a 7B local model
LiteLLM v2 — LLM classifier	6–8/11 (non-deterministic)	paid GPT-5.2 call + ~2–3s on every request; fails outright with local classifier models

The 7 requests LiteLLM v2 misroutes to a 7B ollama model (should be COMPLEX or REASONING):

What you typed	Lynkr sends to	LiteLLM sends to
"Analyse security trade-offs of JWT vs httpOnly cookies for a banking app"	Claude Opus (top)	ollama 7B ❌
"Refactor the entire ingestion pipeline and give me the plan"	z.ai GLM-5.2 (mid)	ollama 7B ❌
"Fix the null-check bug in src/auth/middleware.ts"	Claude Opus (risk-remap)	ollama 7B ❌
"Figure out why this test is flaky. You have full autonomy — iterate until 10 runs pass"	Claude Opus (autonomous)	ollama 7B ❌
"please summarize the exports of this file quickly"	ollama minimax (correct — MEDIUM)	ollama minimax — but for the wrong reason (LiteLLM under-routed to SIMPLE)

Lynkr layers the following, in decreasing priority, to catch every miss LiteLLM leaks:

• FORCE_REASONING regex (ultrathink, prove, security audit, from first principles) → deterministic top-tier
• Risk classifier (auth/middleware/credentials paths) → REASONING under config B
• Agentic detector (AUTONOMOUS workflows) → REASONING minimum
• Anchor-embedding classifier over 4 difficulty classes + FRONTIER_MIN_SIM floor
• Local LLM classifier (qwen2.5:3b) reconciles borderline anchor calls
• Envelope invariance — scores cleaned user text only, ignoring tool schemas / history / system-reminders

LiteLLM's router reads the raw last message with no envelope-stripping, no verify-then-escalate cascade, no risk classifier, no agentic detection — its fallbacks trigger only on HTTP errors, never on a bad answer.

Fairness notes: the 11 scenarios derive from Lynkr's own regression suite, so Lynkr has home-field advantage — the transferable finding is the direction of LiteLLM's failures (systematic under-routing to cheap on hard prompts), not the exact scores. Lynkr's top tiers used Claude Opus 4.8 / z.ai GLM-5.2; LiteLLM was given the identical tier targets.

Cost projection (100,000 requests/month, same backend)

Direct cost (raw meter):

	Monthly cost	vs LiteLLM
LiteLLM (TOON tool-heavy scenario)	~$818	baseline
Lynkr (TOON tool-heavy)	~$409	~50% cheaper via token optimization

Effective cost per correct-tier answer (July 19 head-to-head, all 11 routing scenarios):

	Direct $/mo	Effective $/mo if you re-issue misroutes at correct tier
Lynkr	~$5.50 (spends where it should, doesn't where it shouldn't)	~$5.50
LiteLLM v2 heuristic	$0 (under-routes everything to ollama)	~$89 (7 misroutes need re-issue on Moonshot/COMPLEX)

The story isn't "Lynkr always spends less." It's Lynkr spends where it should and doesn't where it shouldn't. LiteLLM's zero-dollar cost on complex prompts is bought with wrong-tier answers to hard work — a 7B model answering "prove this rate limiter is fair under concurrent refill" is silent quality collapse, not a savings.

→ Full benchmark report with methodology

---

Cost Comparison

Scenario	Direct Anthropic	Lynkr + Ollama	Lynkr + OpenRouter
Daily coding (8h)	$10-30/day	$0 (free)	$2-8/day
Monthly (heavy use)	$300-900	$0	$60-240

With tier routing + token optimization: additional 50-87% savings on cloud providers depending on workload.

---

Why Lynkr vs Alternatives

Feature	Lynkr	LiteLLM	OpenRouter	PortKey
Setup	npm install -g lynkr	Python + Docker + Postgres	Account signup	Docker stack
Claude Code native	✅ Drop-in	⚠️ Requires config	❌	⚠️ Partial
Cursor native	✅ Drop-in	⚠️ Partial	❌	⚠️ Partial
Local models	Ollama, llama.cpp, LM Studio	Ollama only	❌	❌
Automatic tier routing	✅ embedding intent + 13-dimension scorer, verified cascade	⚠️ Auto Router v2 (v1.94): heuristic under-routes, LLM classifier billed per request	❌	❌ Manual metadata
TOON JSON compression	✅ up to 87.6%	❌	❌	❌
Upstream SSE streaming	✅ native passthrough + cross-format transform	⚠️ passthrough only	✅	⚠️
Semantic cache	✅ 171ms hits, 0 tokens	❌	❌	✅ Prompt cache only
Savings & routing dashboard	✅ spend, savings vs flagship, tier mix, routing accuracy, request logs, pivot explorer, session drill-down	⚠️ spend UI only	⚠️ usage page	✅ observability suite (no routing accuracy)
Cost recommendations	✅ evidence-backed findings (dead tool schemas, side-traffic, cache weak spots, Wilson-proven downsizes) ranked by past overspend	❌	❌	⚠️ alerts, no findings
Long-term memory	✅ SQLite, per-session	❌	❌	❌
MCP integration	✅	❌	❌	❌
Self-hosted	✅ Node.js only	✅ Python stack	❌ SaaS	✅ Docker
Dependencies	Node.js 20+	Python, Prisma, PostgreSQL	None	Docker, Python

Lynkr's edge: Purpose-built for AI coding tools. Compresses tokens before they reach the model — not just after. Zero-config for Claude Code, Cursor, and Codex. Installs in one command.

---

Community

• GitHub Discussions — Ask questions
• Report Issues — Bug reports
• NPM Package — Official releases
• DeepWiki — AI-powered docs

This project follows our Code of Conduct — it applies in all project spaces, and it tells you exactly how to report a problem.

---

License

Apache 2.0 — See LICENSE.

---

Built by Vishal Veera Reddy for developers who want control over their AI tools.