Regolo AI
18 modelsProvider docs
As of Oct 4, 2026, Regolo AI lists 18 supported models in the Sovyron catalog, including Qwen3.8 27B, GLM-5.2, Gemma 4 31B IT, Mistral Small 4 119B and GPT OSS 20B — qwen, glm, gemma, mistral-small, gpt-oss and model-router families. Its metered rates span $0.1–$3.2325 per 1M tokens on a 3:1 input:output blend, and every price in the table is Regolo AI's own, normalized to USD — open a model to compare it with the cheapest offer across all providers. Offer mix: 3 free-tier offers.
Supported models and pricing
18 of 18 models
| Model | 1M input | 1M output | 1M cache read | Context | Released | Tags |
|---|---|---|---|---|---|---|
| Qwen3.8 27B | $0.58 | $2.42 | — | 120k | Aug 14, 2026 | |
| GLM-5.2 | $2.31 | $6 | — | 96k | Jun 13, 2026 | |
| Gemma 4 31B IT | $0.46 | $2.42 | — | 100k | Apr 2, 2026 | |
| Mistral Small 4 119B | $0.75 | $3 | — | 256k | Mar 15, 2026 | |
| GPT OSS 20B | $0.4 | $1.8 | — | 128k | Mar 1, 2026 | |
| Qwen Image | $0.5 | $2 | — | 8k | Mar 1, 2026 | |
| Qwen3 Coder Next | $0.3 | $1.2 | — | 262k | Mar 1, 2026 | |
| Brick Complexity Pro | $0.12 | $0.46 | — | 100k | Feb 6, 2026 | |
| Brick v1 Beta | free | free | — | 100k | Feb 6, 2026 | free |
| Qwen3 Embedding 8B | $0.1 | $0.1 | — | 33k | Feb 1, 2026 | |
| Qwen3 Reranker 4B | $0.12 | $0.12 | — | 33k | Feb 1, 2026 | |
| Qwen3.5 9B | $0.15 | $0.6 | — | 262k | Feb 1, 2026 | |
| Qwen3.5-122B | $0.9 | $3.6 | — | 262k | Feb 1, 2026 | |
| DeepSeek OCR 2 | free | free | — | 4k | Jan 27, 2026 | free |
| Apertus 70B | $0.46 | $2.42 | — | 30k | Sep 2, 2025 | |
| GPT OSS 120B | $1 | $4.2 | — | 128k | Aug 5, 2025 | |
| Llama-3.3-70B-Instruct | $0.6 | $2.7 | — | 128k | Apr 28, 2025 | |
| Faster Whisper Large v3 | free | free | — | 448 | Oct 1, 2024 | free |