- Instruction Following
Qwen3.5 Plus 2026-04-20 is a large-scale, proprietary multimodal language model from Qwen (Alibaba) that offers a 1M-token context window and strong reasoning and vision capabilities for…
Powered by Openrouter
Free Models Router
- Instruction Following
Free Models Router is an OpenRouter meta-model that automatically routes requests to compatible free models, providing no-cost inference across multiple underlying LLMs. It filters candidates based on required capabilities such as text, image input, and tool use.
Providers Comparison
Speed
Output tokens per second · Higher is better
L
LLM API
S
SiliconFlow (FP8)
G
GMI
C
CoreWeave
N
Novita
D
DeepSeek
P
Parasail (FP8)
D
DeepInfra (FP4)
End-to-End Response Time
Seconds · Lower is better
L
LLM API
S
SiliconFlow (FP8)
G
GMI
C
CoreWeave
N
Novita
D
DeepSeek
P
Parasail (FP8)
D
DeepInfra (FP4)
Price
USD per 1M tokens (blended) · Lower is better
L
LLM API
S
SiliconFlow (FP8)
G
GMI
C
CoreWeave
N
Novita
D
DeepSeek
P
Parasail (FP8)
D
DeepInfra (FP4)
About the model
What is Free Models Router?
Free Models Router is an OpenRouter routing model (`openrouter/free`) that selects eligible free models to handle each request instead of generating outputs itself. It is mainly used for cost-free experimentation, prototyping, and general text generation across whatever free models are currently available on OpenRouter. It also supports multimodal text-and-image inputs and can be used in applications that require capabilities like vision, reasoning, and tool use without committing to a single backend model. The model belongs to OpenRouter’s router category and is part of its family of meta-models that dynamically dispatch traffic to different hosted LLMs.
Model capabilities
5 Core Capabilities
-
Multi‑model Routing
Routes user requests across multiple free large language models on OpenRouter, selecting an appropriate backend model for each call.
-
General Chat
Supports conversational interactions with natural language understanding and generation, forwarding messages to suitable underlying chat-optimized models.
-
Basic Translation
Relays text translation requests to underlying models that can convert content between multiple languages with reasonable quality.
-
Text From Images
Can pass image inputs to compatible backend models to extract or utilize textual content contained within those images.
-
Image Reasoning
Forwards images to vision-capable models that can interpret visual content, answer questions about images, and describe visual scenes.
Use cases
6 Most Valuable Use Cases
- Cost-Optimized Routing
- Latency-Aware Model Selection
- Automatic Fallback Handling
- Provider Load Balancing
- Usage-Based Model Allocation
- Capability-Based Model Routing
Transparent pricing
Cost Comparison
LLM API offers the lowest effective token costs and best performance among Free Models Router–class APIs.
| Provider | Region | Latency | Throughput | Uptime | Input ($/1M) | Output ($/1M) | Context |
|---|---|---|---|---|---|---|---|
| LLM API BEST | Global | 120ms | 120 tps | 99.99% | $0.00 | $0.00 | 128K |
| Openrouter | Global | ~350ms | ~40 tps | ~99.9% | $0.00 | $0.00 | ~32K |
| OpenAI | Global | ~300ms | ~80 tps | 99.9% | ~$0.10 | ~$0.30 | 128K |
| Anthropic | US East | ~320ms | ~70 tps | 99.9% | ~$0.20 | ~$0.60 | 200K |
| Google AI Studio | Global | ~280ms | ~75 tps | ~99.9% | ~$0.08 | ~$0.24 | ~128K |
Performance benchmarks
Technical Specifications
| Metric | Free Models Router (Openrouter) | OpenAI gpt-4o-mini | Anthropic Claude 3 Haiku |
|---|---|---|---|
| Avg Latency | ~180ms | ~250ms | ~300ms |
| Context Window | ~128K | 128K | 200K |
| Input Price ($/1M) | ~$0.00 | $0.15 | $0.25 |
| Output Price ($/1M) | ~$0.00 | $0.60 | $1.25 |
| Max Output Tokens | ~4K | 4K | 4K |
| Throughput | ~60 tps | ~40 tps | ~35 tps |
| Uptime | ~99.5% | ~99.9% | ~99.9% |
30-day usage via LLM API
- 12.5B
- Prompt tokens processed (30 days)
- 9.1B
- Completion tokens generated (30 days)
- 22.4M
- API requests served (30 days)
- 99.8%
- Avg uptime over last 30 days
Architecture & Integration
Why Build on LLM.API?
One unified API. Every major model. Built-in reliability, cost control, and observability.
-
Unified AI Routing
Dynamically route each request to the best model across providers based on latency, capability, and cost—without changing your integration or redeploying code.
One endpoint, every model -
Cost-Aware Orchestration
Automatically balance quality and spend with configurable cost policies, price-aware routing, and transparent usage data so teams can ship faster without surprise bills.
Optimize quality per dollar -
Resilient Fallback Flows
Define provider-agnostic fallback chains that auto-retry on errors, timeouts, or rate limits to keep your AI features online even when vendors degrade.
Failure-tolerant by design -
End-to-End Observability
Trace every call across providers with logs, metrics, and structured events for prompts, latencies, and errors, making debugging and optimization straightforward in production.
Full visibility into LLMs -
Task-Level Abstractions
Describe what you want—chat, extraction, tools, RAG—and let LLM.API handle provider-specific quirks so teams can swap models without refactoring payloads.
Code to tasks, not vendors -
High-Throughput Batch API
Process large workloads efficiently with batched requests, concurrency controls, and rate-aware scheduling, dramatically reducing per-request overhead and infrastructure complexity.
Scale workloads, not overhead
Decision guide
When to Use — When NOT to Use
Use it if...
- You need a simple way to automatically select among multiple free OpenRouter models.
- You need to prototype quickly without manually benchmarking or hand-picking individual free models.
- Your use case involves non-critical chatbots or helpers where occasional quality variance is acceptable.
- Your use case involves cost-sensitive experimentation and you want to avoid paid models entirely.
- You need a default fallback model when other specific OpenRouter models are unavailable.
Avoid if...
- You need strict, predictable behavior from a single known model for regulatory reasons.
- You need maximum, guaranteed reasoning quality for agents, planning, or complex coding tasks.
- You need stable, reproducible model behavior for benchmarking, research, or A/B experiments.
- Your workload requires finely tuned safety controls and explainable moderation policies per model.
- Your workload requires guaranteed latency, throughput, or availability SLAs from a specific provider.
FAQ
Frequently Asked Questions
-
What is Free Models Router?
Free Models Router is an OpenRouter endpoint that automatically routes your request to one of several free-tier large language models.
-
What is Free Models Router best suited for?
It is best for cost-free experimentation, prototyping, and low-stakes applications where occasional quality or availability variations are acceptable.
-
How is Free Models Router priced when accessed via LLM.API?
Requests through LLM.API are billed according to LLM.API’s pricing for the Free Models Router endpoint, even though the underlying OpenRouter tier is free.
-
What context window does Free Models Router support?
The effective context window depends on the specific underlying free model selected, so you should assume a relatively small to medium context size.
-
How fast is Free Models Router in terms of latency?
Latency can vary per request because different backing models and infrastructures may be selected, so you should not rely on consistent response times.
-
Which modalities does Free Models Router support?
It primarily supports text-in, text-out interactions; image or other modalities are not guaranteed and depend on the routed underlying model.
-
How do I call Free Models Router through the LLM.API gateway?
Use the LLM.API endpoint with the model identifier corresponding to OpenRouter’s Free Models Router and authenticate with your LLM.API key.
-
How does Free Models Router compare to pinned premium models?
Pinned premium models usually provide more predictable quality, latency, and features, while Free Models Router optimizes primarily for zero model-side cost.
-
Are there usage limits or quotas for Free Models Router?
Yes, both OpenRouter’s free tier and LLM.API’s account-level limits can restrict throughput, rate, or total tokens for this model.
-
What are the main limitations of Free Models Router?
You may see inconsistent model behavior, varying capabilities, and occasional capacity errors because requests are routed across multiple free models.
Compare
Competitive Models
- Instruction Following
GPT-5.2 Chat is an OpenAI conversational language model designed for interactive dialogue and task assistance. It focuses on providing coherent, context-aware responses across a wide range…
- Instruction Following
Hailuo 2.3 by MiniMax is a high-fidelity AI video generation model designed for realistic, cinematic 1080p clips from text or image prompts, with strong motion, physics,…