Unverified50% confidenceBenchmarkExact time
In Terminal Bench testing, GPT-6.1 Soul's max/high performance is nearly on par with Opus 5.5, but the cost per task is approximately $0.96 cents, while Opus costs $0.11 to $0.13 per task, roughly one-thirteenth of the price.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/7/2026
First Seen
Valid until: 1/5/2027
Sources
Related Entities
Related Claims
UnverifiedMainstream model inference costs are priced per million tokens: GPT-4o is approximately $5–15, Claude 3.5 Sonnet is approximately $3–15, and the total token count for multi-step agent tasks is often 10–50 times that of a single conversation.73% similarUnverifiedIn the DiD Suite benchmark, the average cost per task at the Max Effort level is as high as $22, while the cost per task at the Low level is only $3.76, a reduction of over 80%.73% similarVerifiedAccounting for real usage scenarios, the actual cost advantage of Opus 5 over Fable is approximately 20% to 25%, with Fable costing about $2.75 per task and Opus 5 about $2.03.73% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/985906API
curl https://kongchang.com/api/v1/knowledge/claims/985906MCP
get_claim(id=985906)