Arcee AI Models — Pricing & Benchmarks
9 models available
Last refreshed 2026-09-30. Next refresh: weekly.
Arcee AI hosts 9 AI models in this catalog. The lowest listed input price is Mistral NeMo Instruct (2407) at $0.15/1M input tokens. LLM Reference compares these models across all 97 providers.
| Model | Input (per 1M) | Output (per 1M) | Context | |
|---|---|---|---|---|
| Mistral NeMo Instruct (2407) | $0.15 | $0.45 | 128k | |
| Gemma 2 27B Instruct | $0.25 | $0.75 | 8k | |
| DeepSeek R1 Distill Llama 70B | $0.35 | $1.05 | 128k | |
| Qwen2.5-Coder-32B-Instruct | $0.4 | $1.20 | 128k | |
| Llama 3.3 70B Instruct (free) | $0.6 | $1.80 | 66k | |
| Trinity Mini | — | — | 128k | |
| Trinity Nano | — | — | 128k | |
| Trinity-Large-Preview | — | — | 128k | |
| Trinity-Large-Thinking | — | — | 256k |
Where else to run this
DeepSeek R1 Distill Llama 70B on Arcee AI
Provider setup and pricing
Qwen2.5-Coder-32B-Instruct on Arcee AI
Provider setup and pricing
Mistral NeMo Instruct (2407) on Arcee AI
Provider setup and pricing
DeepSeek R1 Distill Llama 70B on DeepInfra
Alternative host
Qwen2.5-Coder-32B-Instruct on Cloudflare Workers AI
Alternative host
Mistral NeMo Instruct (2407) on NVIDIA NIM
Alternative host
Pricing Overview
Cheapest$0.15/1M
Most expensive$0.60/1M