General Benchmarks
Conversation understanding and hallucination detection — accuracy vs. average cost per chat.
Reference accuracy vs. cost
Cost — Avg. cost per chat (log scale)
Not published
Aligned benchmark score
Not published
Comparable run cost
Aligned AI coordinates licensed open models through its proprietary orchestration Harness. A reproducible full-Harness evaluation is required before publishing an Aligned score or cost ranking.
Input
$/1M
+
Output
$/1M
External reference costs are estimates based on the listed token pricing and benchmark conversation assumptions. They are not current quotes or measurements of Aligned's operating cost.
This view ranks the external reference entries shown in the scatter plot. Aligned AI is not included without a comparable published Harness result.
| Model | Avg. chat |
|---|---|
Gemini 2.5 Flash Lite | $0.035 Average benchmark cost over 100 conversations, estimated from token pricing. |
GPT 4o Mini | $0.045 Average benchmark cost over 100 conversations, estimated from token pricing. |
Grok 3 Mini | $0.055 Average benchmark cost over 100 conversations, estimated from token pricing. |
Gemini 2 Flash | $0.075 Average benchmark cost over 100 conversations, estimated from token pricing. |
Claude Opus 4.7 | $0.12 Average benchmark cost over 100 conversations, estimated from latest Opus token pricing. |
GPT 5 Mini | $0.15 Average benchmark cost over 100 conversations, estimated from token pricing. |
Gemini 2.5 Pro | $0.21 Average benchmark cost over 100 conversations, estimated from token pricing. |
Gemini 3.1 Pro Preview | $0.29 Average benchmark cost over 100 conversations, estimated from token pricing. |
Claude Opus 4.1 | $0.32 Average benchmark cost over 100 conversations, estimated from token pricing. |
Grok 3 | $0.34 Average benchmark cost over 100 conversations, estimated from token pricing. |
GPT 5.2 | $0.50 Average benchmark cost over 100 conversations, estimated from token pricing. |