Model pricing reference
A complete reference of all models available on Gomus AI, with credit costs per 1,000 tokens.
1 credit = $0.0001 USD. For example, a model costing 30 credits per 1K input tokens means $0.003 per 1K input tokens.
Groq models — Free plan
Available to all users. Powered by Groq's ultra-fast LPU inference.
| Model | Type | Input (per 1K tokens) | Output (per 1K tokens) |
|---|---|---|---|
| Llama 3.3 70B | Chat | 1 | 1 |
| Llama 3.1 8B Instant | Chat | 1 | 1 |
| Llama 4 Maverick | Chat | 1 | 1 |
| Llama 4 Scout | Vision | 1 | 1 |
| Qwen 3 32B | Chat | 1 | 1 |
| Kimi K2 | Chat | 1 | 1 |
| Kimi K2 (0905) | Chat | 1 | 1 |
| GPT OSS 120B | Chat | 1 | 1 |
| GPT OSS 20B | Chat | 1 | 1 |
| Allam 2 7B | Chat | 1 | 1 |
| Groq Compound | Chat | 0 | 0 |
| Groq Compound Mini | Chat | 0 | 0 |
| Llama Guard 4 12B | Safety | 1 | 1 |
| Llama Prompt Guard 2 22M | Safety | 1 | 1 |
| Llama Prompt Guard 2 86M | Safety | 1 | 1 |
| Safety GPT OSS 20B | Safety | 1 | 1 |
| Whisper Large v3 | Speech-to-Text | 1 | 1 |
| Whisper Large v3 Turbo | Speech-to-Text | 1 | 1 |
| Orpheus English | Text-to-Speech | 1 | 1 |
| Orpheus Arabic Saudi | Text-to-Speech | 1 | 1 |
Groq Compound and Compound Mini cost zero credits — ideal for experimenting with agentic workflows at no cost.
AWS Bedrock models — Base plan and above
Require a Base, Premium, or Business subscription. All paid plans have access to the same 22 Bedrock models — the difference between tiers is the monthly credit allowance and resource limits.
Chat & Reasoning
| Model | Input (per 1K tokens) | Output (per 1K tokens) |
|---|---|---|
| Claude Opus 4.6 | 55 | 275 |
| Claude Opus 4.5 | 55 | 275 |
| Claude Sonnet 4.6 | 33 | 165 |
| Claude Sonnet 4.5 | 33 | 165 |
| Claude Sonnet 4 | 30 | 150 |
| Claude 3.7 Sonnet | 30 | 150 |
| Claude 3.5 Sonnet | 30 | 150 |
| Claude 3 Sonnet | 30 | 150 |
| Claude Haiku 4.5 | 11 | 55 |
| Claude 3 Haiku | 3 | 13 |
| Amazon Nova Pro | 11 | 42 |
| Amazon Nova Lite | 1 | 4 |
| Amazon Nova Micro | 1 | 2 |
| Amazon Nova 2 Lite | 5 | 36 |
| Llama 3.2 3B | 2 | 2 |
| Llama 3.2 1B | 2 | 2 |
| Pixtral Large (Mistral) | 20 | 60 |
Embedding
| Model | Input (per 1K tokens) |
|---|---|
| Cohere Embed V4 | 2 |
| Cohere Embed Multilingual V3 | 1 |
| Amazon Titan Embed Text V2 | 2 |
Rerank
| Model | Input (per 1K tokens) |
|---|---|
| Cohere Rerank V3.5 | 20 |
Video
| Model | Input (per 1K tokens) | Output (per 1K tokens) |
|---|---|---|
| TwelveLabs Pegasus V1.2 | 5 | 75 |
Cost examples
To help estimate your usage:
| Scenario | Model | Approx. tokens | Credits used |
|---|---|---|---|
| Quick question | Llama 3.3 70B (Groq) | ~500 in + ~200 out | ~1 |
| Short conversation | Claude 3 Haiku | ~2K in + ~1K out | ~19 |
| Detailed analysis | Claude Sonnet 4 | ~4K in + ~2K out | ~420 |
| Document summary | Amazon Nova Lite | ~10K in + ~2K out | ~18 |
| Knowledge base indexing | Cohere Embed V4 | ~50K in | ~100 |
Token counts are approximate. Actual usage depends on the complexity and length of your prompts and responses.
Subscription plans summary
| Plan | Monthly Credits | Price | Available Models |
|---|---|---|---|
| Free | 1,000 | Free | 20 Groq models |
| Base | 100,000 | $19.90/mo | 20 Groq + 22 Bedrock |
| Premium | 250,000 | $49.90/mo | 20 Groq + 22 Bedrock |
| Business | 750,000 | $149.90/mo | 20 Groq + 22 Bedrock |
For details on how credits work and how to upgrade, see AI Models.