Supported Models
Bear Lumen tracks costs for 31 models across 9 providers. The SDK auto-detects the provider and model from API responses.
Prices last updated .
Token-model rates are per million tokens (USD). Some providers bill per usage unit, so those rows show the per-unit price instead: ElevenLabs per character, Deepgram per minute of audio, and Stability AI per image. Prices are sourced from provider pricing pages and may change. Bear Lumen rate cards are updated regularly.
Anthropic
8 models supported
| Model | Base Name | Input (per 1M tokens) | Output (per 1M tokens) |
|---|---|---|---|
| Claude 3.5 Sonnet | claude-3-5-sonnet | $3.00 | $15.00 |
| Claude 3 Opus | claude-3-opus | $15.00 | $75.00 |
| Claude Haiku 3 | claude-haiku-3 | $0.25 | $1.25 |
| Claude Haiku 4.5 | claude-haiku-4-5 | $1.00 | $5.00 |
| Claude Opus 4 | claude-opus-4 | $15.00 | $75.00 |
| Claude Opus 4.5 | claude-opus-4-5 | $5.00 | $25.00 |
| Claude Sonnet 4 | claude-sonnet-4 | $3.00 | $15.00 |
| Claude Sonnet 4.5 | claude-sonnet-4-5 | $3.00 | $15.00 |
Deepgram
1 models supported
| Model | Base Name | Input (per 1M tokens) | Output (per 1M tokens) |
|---|---|---|---|
| Deepgram Nova 2 | nova-2 | Per unit$0.0043 / minute | |
ElevenLabs
1 models supported
| Model | Base Name | Input (per 1M tokens) | Output (per 1M tokens) |
|---|---|---|---|
| ElevenLabs Multilingual V2 | eleven_multilingual_v2 | Per unit$0.30 / 1K characters | |
7 models supported
| Model | Base Name | Input (per 1M tokens) | Output (per 1M tokens) |
|---|---|---|---|
| Gemini 1.5 Flash | gemini-1.5-flash | $0.075 | $0.30 |
| Gemini 1.5 Pro | gemini-1.5-pro | $1.25 | $5.00 |
| Gemini 2.0 Flash | gemini-2.0-flash | $0.10 | $0.40 |
| Gemini 2.5 Flash | gemini-2.5-flash | $0.30 | $2.50 |
| Gemini 2.5 Flash Lite | gemini-2.5-flash-lite | $0.10 | $0.40 |
| Gemini 2.5 Pro | gemini-2.5-pro | $1.25 | $10.00 |
| Gemini 3 Pro Preview | gemini-3-pro-preview | $2.00 | $12.00 |
Groq
1 models supported
| Model | Base Name | Input (per 1M tokens) | Output (per 1M tokens) |
|---|---|---|---|
| Llama 3.3 70B Versatile | llama-3.3-70b-versatile | $0.59 | $0.79 |
Mistral
2 models supported
| Model | Base Name | Input (per 1M tokens) | Output (per 1M tokens) |
|---|---|---|---|
| Mistral Large | mistral-large | $2.00 | $6.00 |
| Mistral Small | mistral-small | $0.20 | $0.60 |
OpenAI
9 models supported
| Model | Base Name | Input (per 1M tokens) | Output (per 1M tokens) |
|---|---|---|---|
| GPT-3.5 Turbo | gpt-3.5-turbo | $0.50 | $1.50 |
| GPT-4 Turbo | gpt-4-turbo | $10.00 | $30.00 |
| GPT-4.1 | gpt-4.1 | $2.00 | $8.00 |
| GPT-4o | gpt-4o | $2.50 | $10.00 |
| GPT-4o Mini | gpt-4o-mini | $0.15 | $0.60 |
| O1 | o1 | $15.00 | $60.00 |
| O3 | o3 | $2.00 | $8.00 |
| O3 Mini | o3-mini | $1.10 | $4.40 |
| O4 Mini | o4-mini | $1.10 | $4.40 |
Stability AI
1 models supported
| Model | Base Name | Input (per 1M tokens) | Output (per 1M tokens) |
|---|---|---|---|
| Stable Diffusion 3 Medium | sd3-medium | Per unit$0.035 / generation | |
Together AI
1 models supported
| Model | Base Name | Input (per 1M tokens) | Output (per 1M tokens) |
|---|---|---|---|
| Llama 3.3 70B Instruct | meta-llama/Llama-3.3-70B-Instruct | $0.88 | $0.88 |
Don't see your model?
Bear Lumensupports any model with token-based or per-unit pricing. If your provider isn't listed, request support and we'll add it to the rate card catalog. You can also set a custom rate for any model yourself from Model Pricing in your dashboard.
Request model supportHow model detection works
The Bear Lumen SDK inspects the API response to identify the provider and model automatically. Pass the raw provider response to bear.track(response) and costs are calculated using the matching rate card. For models not in the catalog, costs are tracked as zero until a rate card is added.
Set your own pricing
Negotiated a discount with a provider, or running a self-hosted model? Override any model's rate from Model Pricing in your dashboard. Set a custom input and output rate, and Bear Lumenuses it for every cost calculation on that model. Custom rates show a “Custom Rate” badge and take precedence over the public catalog.