Supported Providers
LLM providers that work with Bulk Translate and their configuration details.
Bulk Translate works with any API that exposes an OpenAI-compatible /v1/chat/completions endpoint. This covers virtually every major LLM provider.
Built-in Providers
These eight providers are pre-configured in the Settings dropdown. Selecting one auto-fills the endpoint URL and populates the model list with that provider's popular models.
OpenAI
Endpoint: https://api.openai.com/v1
The default provider. Works with all GPT and o-series models. You'll need an API key from platform.openai.com.
Recommended models: gpt-4o (best overall), gpt-4o-mini (fast and cheap), o3 (reasoning), o4-mini (lightweight reasoning)
Anthropic
Endpoint: https://api.anthropic.com/v1
Claude models excel at nuanced translation and handle context-heavy text well. Get an API key from console.anthropic.com.
Recommended models: claude-sonnet-4-6 (balanced), claude-opus-4-7 (most capable), claude-haiku-4-5 (fastest)
Google Gemini
Endpoint: https://generativelanguage.googleapis.com/v1beta
Gemini offers strong multilingual performance with competitive pricing. API keys available at aistudio.google.com.
Recommended models: gemini-2.5-pro (best quality), gemini-2.5-flash (speed), gemini-2.0-flash (budget)
Groq
Endpoint: https://api.groq.com/openai/v1
Groq runs open-weight models on custom hardware, delivering the fastest inference speeds — great for large batches. Sign up at console.groq.com.
Recommended models: llama-4-maverick-17b-128e-instruct (most capable), deepseek-r1-distill-llama-70b (reasoning), qwen-2.5-32b (balanced)
DeepSeek
Endpoint: https://api.deepseek.com/v1
DeepSeek models offer strong translation quality at very low prices. Register at platform.deepseek.com.
Recommended models: deepseek-chat (general purpose), deepseek-reasoner (complex text)
OpenRouter
Endpoint: https://openrouter.ai/api/v1
OpenRouter is a unified API that gives you access to 200+ models from every major provider through a single key. Useful for comparing models or accessing models you don't have direct API access to. Sign up at openrouter.ai.
Recommended models: openai/gpt-4o, anthropic/claude-sonnet-4-6, google/gemini-2.5-pro
Together AI
Endpoint: https://api.together.xyz/v1
Together AI hosts open-weight models with straightforward pricing. Get started at together.ai.
Recommended models: meta-llama/Llama-4-Maverick-17B-128E-Instruct, deepseek-ai/DeepSeek-R1, Qwen/Qwen2.5-72B-Instruct
Ollama (Local)
Endpoint: http://localhost:11434/v1
Run models entirely on your machine — zero API cost, complete privacy. Install Ollama, pull a model (ollama pull llama3.2), and select "Ollama" in the provider dropdown.
Recommended models: llama3.2, mistral, gemma3, codellama
Custom Endpoints
If your provider isn't listed, select "Custom endpoint…" from the provider dropdown. Enter any OpenAI-compatible base URL, and the model field becomes a free-text input.
This works with:
- Self-hosted solutions like vLLM, LocalAI, or text-generation-webui
- Regional providers like Mistral, Cohere, or AI21
- Proxy services like LiteLLM or a custom gateway
- Any endpoint that speaks the OpenAI chat completions protocol
Pricing & Rate Limits
Bulk Translate doesn't add any markup. You pay your provider directly at their published rates. Check your provider's dashboard for usage, billing, and rate limit information.
Tip: Start with a cheaper model (like gpt-4o-mini or gemini-2.0-flash) for testing, then switch to a more capable model when you need higher quality.