# ClaimBench / subtle contradiction

> Source: Benchize (https://benchize.com) — https://benchize.com/benchmarks/vals-ai%3Aclaimbench%3Asubtle_contradiction. When citing or reusing this content, credit "Benchize" and link to https://benchize.com/benchmarks/vals-ai%3Aclaimbench%3Asubtle_contradiction.

Does a 10-K excerpt support a claim, contradict it, or say nothing about it?

Canonical URL: https://benchize.com/benchmarks/vals-ai%3Aclaimbench%3Asubtle_contradiction
Last updated: Wed, 21 Oct 2026 18:52:20 GMT

Reproduction and quotation must name Benchize, link to the canonical HTML URL, and retain individual source citations.

Conquered

Does a 10-K excerpt support a claim, contradict it, or say nothing about it?

Professional work · Models tested: 5 · Score publishers: Vals AI

## Model results

| # | Model | Score | ± | Cost | Latency |  |
| --- | --- | --- | --- | --- | --- | --- |
| 1 | [Claude Haiku 5.5](https://benchize.com/models/claude-haiku-5-5-max.md)Max | 100% | SE ±0 pp | $0.00251 | 5.32source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 100 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.002508 \| USD \| per_test \| Lower is better \| \| `latency` \| 5.318388 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature Not reported topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0 pp sourcePopulation 19 stderr 0 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_claimbench.cK2bPt2E.json) Source Vals AI Original model label anthropic/claude-haiku-5-5 Source version 1 Observed 2026-10-21T16:45:56.452Z Evidence SHA-256 `c0b96e43a97ed7a9dbb35a06b2ff59446cf8fd70f6d50005b8c58aa860f229d8` Source locator `/tasks/subtle_contradiction/anthropic~1claude-haiku-5-5` |  |  |  |  |  |  |
| 1 | [Claude Sonnet 5.5](https://benchize.com/models/claude-sonnet-5-5-max.md)Max | 100% | SE ±0 pp | $0.04 | 4.05source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 100 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.044616 \| USD \| per_test \| Lower is better \| \| `latency` \| 4.048482 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0 pp sourcePopulation 19 stderr 0 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_claimbench.cK2bPt2E.json) Source Vals AI Original model label anthropic/claude-sonnet-5-5 Source version 1 Observed 2026-10-21T16:45:56.452Z Evidence SHA-256 `c0b96e43a97ed7a9dbb35a06b2ff59446cf8fd70f6d50005b8c58aa860f229d8` Source locator `/tasks/subtle_contradiction/anthropic~1claude-sonnet-5-5` |  |  |  |  |  |  |
| 1 | [Claude Opus 5.5](https://benchize.com/models/claude-opus-5-5-max.md)Max | 100% | SE ±0 pp | $0.09 | 6.19source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 100 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.088483 \| USD \| per_test \| Lower is better \| \| `latency` \| 6.187811 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0 pp sourcePopulation 19 stderr 0 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_claimbench.cK2bPt2E.json) Source Vals AI Original model label anthropic/claude-opus-5-5 Source version 1 Observed 2026-10-21T16:45:56.452Z Evidence SHA-256 `c0b96e43a97ed7a9dbb35a06b2ff59446cf8fd70f6d50005b8c58aa860f229d8` Source locator `/tasks/subtle_contradiction/anthropic~1claude-opus-5-5` |  |  |  |  |  |  |
| 1 | [GPT-6.1 Sol](https://benchize.com/models/gpt-6-1-sol-max.md)Max | 100% | SE ±0 pp | $0.03 | 10.91source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 100 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.028341 \| USD \| per_test \| Lower is better \| \| `latency` \| 10.905898 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 128000 provider OpenAI reasoning Not reported Reasoning effort max Temperature Not reported topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0 pp sourcePopulation 19 stderr 0 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_claimbench.cK2bPt2E.json) Source Vals AI Original model label openai/gpt-6.1-sol Source version 1 Observed 2026-10-21T16:45:56.452Z Evidence SHA-256 `c0b96e43a97ed7a9dbb35a06b2ff59446cf8fd70f6d50005b8c58aa860f229d8` Source locator `/tasks/subtle_contradiction/openai~1gpt-6.1-sol` |  |  |  |  |  |  |
| 5 | [GPT-6 Luna](https://benchize.com/models/gpt-6-luna-max.md)Max | 95.92% | SE ±1 pp | $0.00136 | 7.87source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 95.918 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.001357 \| USD \| per_test \| Lower is better \| \| `latency` \| 7.866128 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 128000 provider OpenAI reasoning Not reported Reasoning effort max Temperature Not reported topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1 pp sourcePopulation 19 stderr 1.001 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_claimbench.cK2bPt2E.json) Source Vals AI Original model label openai/gpt-6-luna Source version 1 Observed 2026-10-21T16:45:56.452Z Evidence SHA-256 `c0b96e43a97ed7a9dbb35a06b2ff59446cf8fd70f6d50005b8c58aa860f229d8` Source locator `/tasks/subtle_contradiction/openai~1gpt-6-luna` |  |  |  |  |  |  |

---
Cite as: ClaimBench / subtle contradiction — Benchize, https://benchize.com/benchmarks/vals-ai%3Aclaimbench%3Asubtle_contradiction (retrieved 2026-10-21)