Supported Models
evroc offers a curated catalogue of shared models across text generation, reasoning, coding, multimodal, embedding, reranking, and audio transcription. Each card below lists the model identifier, context and output limits, supported capabilities, and pricing per million tokens (input / output). Pricing is in EUR and fetched live from the evroc billing API.
zai-org/GLM-5.2
Open flagship GLM for long-horizon coding agents and million-token context work.
google/gemma-4-26B-A4B-it
Open Gemma instruction model for efficient chat and self-hosted deployments.
openai/gpt-oss-120b
Open GPT reasoning model for self-hosted agents and controllable deployments.
moonshotai/Kimi-K2.6
Multimodal Kimi workhorse for agent loops, coding tasks, and visual context.
nvidia/Llama-3.3-70B-Instruct-FP8
Popular open Llama workhorse for multilingual chat, coding, and self-hosting.
mistralai/Mistral-Medium-3.5-128B
Balanced Mistral model for enterprise assistants, multilingual work, and tools.
Qwen/Qwen3.6-35B-A3B
Open multimodal Qwen MoE for local agents that need vision, audio, and code.
Qwen/Qwen3.8-27B
Dense 27B vision-language model for coding, agent tasks, and image and video understanding.
evroc/roc
evroc's high-intelligence model for complex reasoning, coding, and agentic work.
intfloat/multilingual-e5-large-instruct
Multilingual embedding model for semantic search and retrieval across 100+ languages.
Qwen/Qwen3-Embedding-8B
High-performance text embedding model for semantic search, retrieval, and clustering.
Qwen/Qwen3-Reranker-4B
Lightweight reranker for improving search relevance by reordering retrieved documents.
mistralai/Voxtral-Small-24B-2507
Multimodal Mistral model for spoken-language understanding and audio transcription.
KBLab/kb-whisper-large
Whisper-based Swedish transcription model trained by KBLab for high-quality captioning.
openai/whisper-large-v3
Open Whisper checkpoint for robust multilingual transcription and captioning.
openai/whisper-large-v3-turbo
Speech transcription model for accurate audio-to-text and captioning workflows.
See Also
- evroc Think Models concepts — Core concepts: models, instances, and shared models
- evroc Think Models CLI — Manage model instances and API keys from the command line
- Inference API — OpenAI-compatible API reference