GreatRouter charges per request based on the model used. Every price is denominated with the correct unit — no hidden fees, no confusing line items.
How It Works
Pay-Per-Use
You are charged only for the requests you make. No minimums, no monthly commitments, no surprise fees.
Auto-Recharge
When your credit balance drops below a configurable threshold, your payment method is automatically charged a predefined amount. You control both the threshold and the recharge amount.
Built-In Logging & Tracing
Every request is logged with full visibility: model used, token count, latency, cost, and request/response data. Access detailed traces from your dashboard.
Real-Time Cost Tracking
See your spending by model, provider, and time period — updated in real time.
Text Generation Pricing
Text models are billed per million tokens (input and output separately).
| Model | Input | Output | Cached Input |
|---|
| GPT-5 (OpenAI) | $1.25 / 1M tokens | $10.00 / 1M tokens | $0.125 / 1M tokens |
| GPT-5.4 (OpenAI) | $2.50 / 1M tokens | $15.00 / 1M tokens | $0.25 / 1M tokens |
| GPT-5.4 Mini (OpenAI) | $0.40 / 1M tokens | $1.60 / 1M tokens | — |
| GPT-5.4 Nano (OpenAI) | $0.20 / 1M tokens | $0.80 / 1M tokens | — |
| GPT-5.4 Pro (OpenAI) | $30.00 / 1M tokens | $60.00 / 1M tokens | — |
| GPT-5.5 (OpenAI) | $5.00 / 1M tokens | $25.00 / 1M tokens | — |
| GPT-5.5 Pro (OpenAI) | $30.00 / 1M tokens | $150.00 / 1M tokens | — |
| o4 Mini (OpenAI) | $1.10 / 1M tokens | $4.40 / 1M tokens | — |
| Claude Sonnet 4 (Anthropic) | $3.00 / 1M tokens | $15.00 / 1M tokens | $0.30 / 1M tokens |
| Claude Sonnet 4.5 (Anthropic) | $3.00 / 1M tokens | $15.00 / 1M tokens | $0.30 / 1M tokens |
| Claude Sonnet 4.6 (Anthropic) | $3.00 / 1M tokens | $15.00 / 1M tokens | $0.30 / 1M tokens |
| Claude Haiku 4.5 (Anthropic) | $1.00 / 1M tokens | $5.00 / 1M tokens | $0.10 / 1M tokens |
| Claude Opus 4.6 (Anthropic) | $5.00 / 1M tokens | $25.00 / 1M tokens | $0.50 / 1M tokens |
| Claude Opus 4.7 (Anthropic) | $5.00 / 1M tokens | $25.00 / 1M tokens | $0.50 / 1M tokens |
| Claude Opus 4.8 (Anthropic) | $5.00 / 1M tokens | $25.00 / 1M tokens | $0.50 / 1M tokens |
| Gemini 2.5 Flash (Google) | $0.30 / 1M tokens | $2.50 / 1M tokens | $0.075 / 1M tokens |
| Gemini 2.5 Flash Lite (Google) | $0.10 / 1M tokens | $0.40 / 1M tokens | — |
| Gemini 2.5 Pro (Google) | $1.25 / 1M tokens | $10.00 / 1M tokens | $0.3125 / 1M tokens |
| Gemini 3 Flash (Google) | $0.50 / 1M tokens | $3.00 / 1M tokens | $0.05 / 1M tokens |
| Gemini 3.1 Flash Lite (Google) | $0.25 / 1M tokens | $1.00 / 1M tokens | — |
| Gemini 3.1 Pro (Google) | $1.25 / 1M tokens | $5.00 / 1M tokens | $0.3125 / 1M tokens |
| Qwen3 Max (Alibaba) | $1.25 / 1M tokens | $5.00 / 1M tokens | — |
| Qwen3.5 397B A17B (Alibaba) | $1.50 / 1M tokens | $6.00 / 1M tokens | — |
| LLaMA 3.1 70B Instruct (Meta) | $0.293 / 1M tokens | $2.253 / 1M tokens | — |
| LLaMA 3.3 70B (Meta) | $0.293 / 1M tokens | $2.253 / 1M tokens | — |
| LLaMA 4 Scout (Meta) | $0.27 / 1M tokens | $0.85 / 1M tokens | — |
| Mistral Small 3.1 24B (Mistral AI) | $0.351 / 1M tokens | $0.555 / 1M tokens | — |
| GEMA-SEA-LION v4 27B (AI Singapore) | $0.351 / 1M tokens | $0.555 / 1M tokens | — |
| Kimi K2.5 (Moonshot AI) | $0.60 / 1M tokens | $3.00 / 1M tokens | — |
| Kimi K2.6 (Moonshot AI) | $0.95 / 1M tokens | $4.75 / 1M tokens | — |
| GLM 4.7 Flash (Z.ai) | $0.061 / 1M tokens | $0.40 / 1M tokens | — |
| DeepSeek R1 Distill 32B (DeepSeek) | $0.497 / 1M tokens | $4.881 / 1M tokens | — |
| Grok 4.3 (xAI) | $1.25 / 1M tokens | $2.50 / 1M tokens | $0.20 / 1M tokens |
| Grok 4.20 Reasoning (xAI) | $2.00 / 1M tokens | $10.00 / 1M tokens | — |
| Grok 4.20 Non-Reasoning (xAI) | $2.00 / 1M tokens | $10.00 / 1M tokens | — |
| Grok 4.20 Multi-Agent (xAI) | $2.00 / 1M tokens | $10.00 / 1M tokens | — |
| Nemotron 3 120B (NVIDIA) | $0.50 / 1M tokens | $2.50 / 1M tokens | — |
| GPT-OSS 120B (OpenAI) | $0.35 / 1M tokens | $1.40 / 1M tokens | — |
| GPT-OSS 20B (OpenAI) | $0.20 / 1M tokens | $0.80 / 1M tokens | — |
| M2.7 (MiniMax) | $0.30 / 1M tokens | $1.50 / 1M tokens | — |
| Qwen2.5 Coder 32B (Qwen) | $0.66 / 1M tokens | $2.64 / 1M tokens | — |
| QwQ 32B (Qwen) | $0.66 / 1M tokens | $2.64 / 1M tokens | — |
| Qwen3 30B A3B FP8 (Qwen) | $0.0509 / 1M tokens | $0.0509 / 1M tokens | — |
LoRA / Fine-Tunable Base Models
These models support LoRA fine-tuning. Pricing is per-adapter when fine-tuned; base model pricing shown.
| Model | Input | Output |
|---|
| Gemma 2B IT LoRA (Google) | $0.01 / 1M tokens | — |
| Gemma 7B IT (Google) | $0.05 / 1M tokens | $0.05 / 1M tokens |
| Gemma 7B IT LoRA (Google) | $0.05 / 1M tokens | — |
| Llama 2 7B Chat HF LoRA (Meta) | $0.06 / 1M tokens | — |
| Llama 3.1 70B Instruct (Meta) | $0.01 / 1M tokens | $0.01 / 1M tokens |
| Mistral 7B Instruct v0.2 | $0.05 / 1M tokens | $0.05 / 1M tokens |
| Mistral 7B Instruct v0.2 LoRA | $0.05 / 1M tokens | — |
Image Generation Pricing
Image models are billed per image, with pricing that varies by resolution and model.
Premium Image Models
| Model | Price |
|---|
| Flux 1 Schnell (Black Forest Labs) | $0.000053 / image |
| Flux 2 Dev (Black Forest Labs) | tile-step pricing from $0.00021 |
| Flux 2 Flex (Black Forest Labs) | $0.05 / MP output, $0.05 / MP input |
| Flux 2 Klein 4B (Black Forest Labs) | $0.000059 / input tile, $0.000287 / output tile |
| Flux 2 Klein 9B (Black Forest Labs) | $0.015 / image |
| Flux 2 Max (Black Forest Labs) | $0.07 / 1st MP, $0.03 / addtl MP, $0.03 / input MP |
| Flux 2 Pro Preview (Black Forest Labs) | $0.03 / 1st MP, $0.015 / addtl MP, $0.015 / input MP |
| Recraft V4 (Recraft) | $0.04 / image |
| Recraft V4 Pro (Recraft) | $0.25 / image |
| Recraft V4 Vector (Recraft) | $0.08 / image |
| Recraft V4 Pro Vector (Recraft) | $0.30 / image |
| Recraft V4.1 (Recraft) | $0.04 / image |
| Recraft V4.1 Utility (Recraft) | $0.04 / image |
| Recraft V4.1 Utility Pro (Recraft) | $0.25 / image |
| Recraft V4.1 Pro (Recraft) | $0.25 / image |
| Recraft V4.1 Vector (Recraft) | $0.08 / image |
| Recraft V4.1 Pro Vector (Recraft) | $0.30 / image |
| Recraft V4.1 Utility Vector (Recraft) | $0.08 / image |
| Recraft V4.1 Utility Pro Vector (Recraft) | $0.30 / image |
| Seedream 4.0 (ByteDance) | $0.03 / image |
| Seedream 4.5 (ByteDance) | $0.04 / image |
| Seedream 5 Lite (ByteDance) | $0.035 / image |
| Seedance 2.0 (ByteDance) | see video pricing |
| Imagen 4 (Google) | $0.04 / image |
| Nano Banana (Google) | input $0.30 / 1M tokens, output $30.00 / 1M tokens |
| Nano Banana 2 (Google) | input $0.50 / 1M tokens, output $60.00 / 1M tokens |
| Nano Banana Pro (Google) | input $2.00 / 1M tokens, output $120.00 / 1M tokens |
| Grok Imagine (xAI) | $0.02 / image |
| Grok Imagine Quality (xAI) | $0.05 / image |
| GPT Image 1.5 (OpenAI) | input $5.00 / 1M tokens, input images $8.00 / 1M tokens, output $10.00 / 1M tokens |
| GPT Image 2 (OpenAI) | input $5.00 / 1M tokens, input images $8.00 / 1M tokens, output $10.00 / 1M tokens |
| Wan 2.6 Image (Alibaba) | $0.03 / image |
| Phoenix 1.0 (Leonardo) | $0.006 / image |
| Lucid Origin (Leonardo) | $0.007 / image |
Standard Image Models
| Model | Price |
|---|
| Stable Diffusion XL Lightning (ByteDance) | $0.035 / image |
| Dreamshaper 8 LCM (Lykon) | $0.035 / image |
| Stable Diffusion v1.5 Img2Img (RunwayML) | $0.035 / image |
| Stable Diffusion v1.5 Inpainting (RunwayML) | $0.035 / image |
| Stable Diffusion XL Base 1.0 (Stability AI) | $0.035 / image |
Video Generation Pricing
Video models are billed per second of output.
| Model | Price |
|---|
| Veo 3 Fast (Google) | $0.08 / sec (720p), $0.10 / sec (1080p), $0.10 / sec (720p w/ audio), $0.12 / sec (1080p w/ audio), $0.30 / sec (4K w/ audio) |
| Veo 3 (Google) | $0.20 / sec (720p), $0.20 / sec (1080p), $0.40 / sec (720p w/ audio), $0.40 / sec (1080p w/ audio) |
| Veo 3.1 Fast (Google) | $0.08 / sec (720p), $0.10 / sec (1080p), $0.25 / sec (4K), $0.10 / sec (720p w/ audio), $0.12 / sec (1080p w/ audio), $0.30 / sec (4K w/ audio) |
| Veo 3.1 (Google) | $0.20 / sec (720p), $0.20 / sec (1080p), $0.40 / sec (4K), $0.40 / sec (720p w/ audio), $0.40 / sec (1080p w/ audio), $0.60 / sec (4K w/ audio) |
| PixVerse V6 | $0.025 / sec (360p), $0.035 / sec (540p), $0.045 / sec (720p), $0.090 / sec (1080p) |
| PixVerse V5.6 | tiered by resolution + duration |
| Vidu Q3 Turbo | $0.04 / sec (540p), $0.06 / sec (720p), $0.07 / sec (1080p) |
| Vidu Q3 Pro | $0.05 / sec (540p), $0.125 / sec (720p), $0.15 / sec (1080p) |
| Hailuo 2.3 (MiniMax) | $0.047 / sec |
| Hailuo 2.3 Fast (MiniMax) | $0.032 / sec |
| Seedance 2.0 (ByteDance) | $0.22 / sec (720p), $0.55 / sec (1080p) |
| Seedance 2.0 Fast (ByteDance) | $0.08 / sec (720p), $0.17 / sec (1080p) |
| HH1-T2V (Alibaba) | $0.14 / sec (720p), $0.28 / sec (1080p) |
| HH1-I2V (Alibaba) | $0.14 / sec (720p), $0.28 / sec (1080p) |
| Gen 4.5 (Runway) | $0.12 / sec |
| Grok Imagine Video (xAI) | $0.05 / sec |
| Grok Imagine Video 1.5 Preview (xAI) | $0.08 / sec, $0.14 / sec (720p) |
Image-to-Video Models
| Model | Price |
|---|
| Wan 2.7 I2V (Alibaba) | $0.10 / sec (720p), $0.15 / sec (1080p) |
Audio Pricing
Speech-to-Text
| Model | Price |
|---|
| Deepgram Nova-3 | $0.0052 / 1M input tokens |
| Deepgram Flux | $0.0077 / 1M input tokens |
| Whisper (OpenAI) | $0.000453 / 1M input tokens |
| Whisper Tiny EN (OpenAI) | $0.000453 / 1M input tokens |
| Whisper Large V3 Turbo (OpenAI) | $0.000513 / 1M input tokens |
| Universal 3 Pro (AssemblyAI) | $0.0035 / audio minute |
| GPT-4o Transcribe (OpenAI) | $0.006 / audio minute |
| Grok STT (xAI) | $0.001667 / audio minute |
Text-to-Speech
| Model | Price |
|---|
| TTS-1 (OpenAI) | $0.000015 / character |
| TTS-1 HD (OpenAI) | $0.00003 / character |
| Aura 1 (Deepgram) | $0.015 / 1M input tokens |
| Aura 2 English (Deepgram) | $0.03 / 1M input tokens |
| Aura 2 Spanish (Deepgram) | $0.03 / 1M input tokens |
| Speech 2.8 Turbo (MiniMax) | $0.00006 / character |
| Speech 2.8 HD (MiniMax) | $0.0001 / character |
| TTS 2 (Inworld) | $0.000035 / character |
| TTS 1.5 Max (Inworld) | $0.000035 / character |
| TTS 1.5 Mini (Inworld) | $0.000025 / character |
| MeloTTS (MyShell) | $0.000205 / 1M input tokens |
| Grok TTS (xAI) | $0.000015 / character |
Gemini 3.1 Flash TTS Pricing
| Token Type | Price |
|---|
| Input text tokens | $0.75 / 1M tokens |
| Input audio tokens | $3.00 / 1M tokens |
| Output text tokens | $4.50 / 1M tokens |
| Output audio tokens | $12.00 / 1M tokens |
Classification Pricing
| Model | Price |
|---|
| ResNet-50 (Microsoft) | $0.00000251 / 1M input tokens |
| DistilBERT SST-2 (Hugging Face) | $0.0263 / 1M input tokens |
| BGE Reranker Base (BAAI) | $0.00311 / 1M input tokens |
Music Generation
Music and audio models bill per generation, track, request, or compute second depending on the provider. See the Models catalog for the full routable list with units.
| Model | Price | Unit |
|---|
| music generator (CassetteAI) | $0.0013 | per compute second |
| Elevenlabs Music | $0.80 | per generation |
| Music 2.6 (MiniMax) | $0.01 / $0.15 | per track (lyrics / full) |
| Minimax Music 2.5 / 2.6 | $0.15 | per generation |
| Suno V4.5 / V5 | varies | per generation (async) |
Translation Pricing
| Model | Price |
|---|
| M2M100 1.2B (Meta) | $0.342 / 1M tokens (input + output) |
| IndicTrans2 (AI4Bharat) | $0.342 / 1M tokens (input + output) |
Embeddings Pricing
| Model | Price |
|---|
| BGE Small EN v1.5 (BAAI) | $0.0202 / 1M tokens |
| BGE Base EN v1.5 (BAAI) | $0.0666 / 1M tokens |
| BGE Large EN v1.5 (BAAI) | $0.204 / 1M tokens |
| BGE M3 (BAAI) | $0.0118 / 1M tokens |
| EmbeddingGemma 300M (Google) | $0.01 / 1M tokens |
| PLaMo Embedding 1B (Preferred Networks) | $0.0186 / 1M tokens |
| Qwen3 Embedding 0.6B (Qwen) | $0.0118 / 1M tokens |
Image-to-Text Pricing
| Model | Price |
|---|
| LLaVA 1.5 7B (LLaVA-HF) | $0.05 / 1M input tokens, $0.05 / 1M output tokens |
Voice Activity Detection
| Model | Price |
|---|
| Smart Turn V2 (Pipecat AI) | $0.000338 / 1M input tokens |
Cost Estimation Examples
Summarize a 10-page PDF
- Input: ~3,000 tokens
- Model: LLaMA 3.3 70B (via auto-route)
- Cost: ~$0.0009 (3K input × $0.293 / 1M + ~1K output × $2.253 / 1M)
Generate an image
- Model: Flux 2 Flex (via auto-route)
- Cost: ~$0.05
Transcribe a 30-minute meeting
- Model: Deepgram Nova-3 (via auto-route)
- Cost: ~$0.156 (30 min × $0.0052 / 1M tokens)
Generate 10 seconds of 720p video
- Model: Veo 3 Fast (via auto-route)
- Cost: ~$0.80 (10 × $0.08)
Price Tiers
Every model is assigned a price tier to help you quickly compare cost across models. Tiers use percentile ranking within each category — a model’s tier depends on how it compares to other models that do the same job.
| Tier | Meaning | Percentile |
|---|
| Economy | Lowest-cost options in the category | 0–20th |
| Standard | Below-average cost, good value | 20–40th |
| Balanced | Near the category average | 40–60th |
| Premium | Above-average cost, higher capability | 60–80th |
| Flagship | Top-tier, most capable and expensive | 80–100th |
How it works
- Models are grouped by category (LLM, image generation, video generation, etc.) and sorted by average price.
- Each tier holds roughly 20% of models in that category.
- Balanced means near the median price for that category.
- Tiers are relative within categories — an Economy LLM may cost more in absolute dollars than a Flagship TTS model.
- Categories with fewer than 5 models use a 3-tier split (Economy / Balanced / Flagship); single-model categories are always Balanced.
You can use the maxCost parameter on the /v1/models/route endpoint to filter out models above a given tier, or filter by price_tier in model listings.
Fair Usage
GreatRouter has no minimum commitment. You pay only for the requests you make. Free credits are included with every new account for testing and evaluation.