Aion-RP-Llama-3.1-8B ranks the highest in the character evaluation portion of the RPBench-Auto benchmark, a roleplaying-specific variant of Arena-Hard-Auto, where LLMs evaluate each other’s responses. It is a fine-tuned base model...
Model details →AionLabs: Aion-RP 1.0 (8B) vs Magnum v4 72B
AionLabs: Aion-RP 1.0 (8B) wins on 2 of 6 axes — pricing and capability skew in its favour for most workloads.
This is a series of models designed to replicate the prose quality of the Claude 3 models, specifically Sonnet(https://openrouter.ai/anthropic/claude-3.5-sonnet) and Opus(https://openrouter.ai/anthropic/claude-3-opus). The model is fine-tu…
Model details →Side-by-side comparison
| Capability | AionLabs: Aion-RP 1.0 (8B) | Magnum v4 72B | Winner |
|---|---|---|---|
| Context window Maximum number of input tokens the model can attend to in a single request. | 33K | 33K | Tie |
| Input price (per 1M) Cost per million input tokens billed by the provider. | $0.80 | $2.50 | 🏆 AionLabs: Aion-RP 1.0 (8B) |
| Output price (per 1M) Cost per million output tokens billed by the provider. | $1.60 | $5.00 | 🏆 AionLabs: Aion-RP 1.0 (8B) |
| Tool / function calling First-class support for emitting structured tool calls. | — | — | Tie |
| Vision input Accepts image inputs alongside text. | — | — | Tie |
| Reasoning mode Internal chain-of-thought / extended-thinking support. | — | — | Tie |
Frequently asked questions
Is AionLabs: Aion-RP 1.0 (8B) better than Magnum v4 72B?
AionLabs: Aion-RP 1.0 (8B) wins on 2 of 6 axes — pricing and capability skew in its favour for most workloads.
What's the price difference between AionLabs: Aion-RP 1.0 (8B) and Magnum v4 72B?
Input: $0.80 vs $2.50 per 1M tokens. Output: $1.60 vs $5.00 per 1M tokens.
What context windows do AionLabs: Aion-RP 1.0 (8B) and Magnum v4 72B support?
AionLabs: Aion-RP 1.0 (8B) supports up to 33K tokens. Magnum v4 72B supports up to 33K tokens.
Do both AionLabs: Aion-RP 1.0 (8B) and Magnum v4 72B support tool calling?
AionLabs: Aion-RP 1.0 (8B): not advertised. Magnum v4 72B: not advertised.