INDEPENDENT MEASUREMENT • PLAIN-LANGUAGE READOUT • UPDATED DAILY

TAB Daily Results

2026-11-10

Deviation from baseline detected in 14 cells across 14 models this cycle — 6 above baseline, 8 below baseline. Detail below.
Interpretation notice — rule_based_provenance v1.0.0. Citation-related checks use deterministic citation-shape detection and comparison against case-supplied source material or allowed-source sets. This instrument does not resolve URLs or DOIs, search the open web, or independently verify the real-world existence or meaning of a cited source. A failing result records only the mechanical observation the scorer describes; it does not, by itself, determine whether a cited source genuinely exists, nor determine model intent. Source admission is a closed set of three authored cases, so a cell takes one of four values and one case moves it by a third of the subscale; it is a bounded subscale score on those three prompts, not a rate over general use and not an estimate of how often a model would do this elsewhere. In the September 3-4 2026 current-regime window - 38 model-by-cycle cells, three repeats each - repeated measurement of the same model differed by one of the three cases more often than not, so within that window a single-cycle difference of one or two cases is not a resolved difference between models. That estimate is one short window and is not a general constant. Interpreting differences across repeated cycles or by case recurrence requires a pre-specified rule, which has not yet been established. Where a HOLD is recorded, it withholds an acknowledgment pending review; it is a fail-closed measurement control and is not itself evidence about a model. Historical scores are unchanged; this notice clarifies the claim boundary supported by those scores.
Measurement validity notice, September 1, 2026. The current flattened Instruction Authority instrument is frozen for aggregate directional interpretation. Its 15 scenarios are delivered through one user-role message while authored authority tiers are described in-band, and subsequent raw-response review identified construct/scorer-direction contradictions. Historical observations remain unchanged. The aggregate combines multiple constructs and cannot support a single better/worse interpretation pending decomposition and successor-instrument validation.

By model

What today’s completed cycle measured, one model at a time. Every line is a fact from the same data behind the Index.

claude-haiku-4-5

claude-haiku-4-5 scored 15/15 on provenance, tied for the category lead.

deepseek-v3.2

deepseek-v3.2 scored 14/14 on trust and reliability, tied for the category lead.

deepseek-v3.2 scored 9/15 on provenance, within repeat variation vs prior run 11/15, lowest in field.

Prior run: 11/15. 7 days ago: 9/15. 30 days ago: 9/15.

3 results contained an uncertainty admission plus a citation-shaped token matching no allowed source.

glm-5.2

glm-5.2 scored 14/14 on trust and reliability, tied for the category lead.

glm-5.2 scored 14/15 on provenance, within repeat variation vs prior run 13/15.

Prior run: 13/15. 7 days ago: 15/15. 30 days ago: 14/15.

1 result contained an uncertainty admission plus a citation-shaped token matching no allowed source.

glm-5.2 scored 15/15 on instruction authority, tied for the category lead.

claude-fable-5

claude-fable-5 scored 14/14 on trust and reliability, tied for the category lead.

claude-fable-5 scored 12/15 on provenance.

Prior run: 12/15. 7 days ago: 12/15. 30 days ago: n/a (insufficient comparable history).

1 result contained an uncertainty admission plus a citation-shaped token matching no allowed source.

claude-fable-5 engaged with 14 of 15 instruction authority cases, 1 refused. Prior run: 14 of 15. 7 days ago: 14 of 15. 30 days ago: n/a (insufficient comparable history).

grok-4.5

grok-4.5 scored 14/14 on trust and reliability, tied for the category lead.

grok-4.5 scored 15/15 on provenance, tied for the category lead.

Prior run: 15/15. 7 days ago: 15/15. 30 days ago: n/a (insufficient comparable history).

grok-4.5 scored 15/15 on instruction authority, tied for the category lead.

gpt-5.6-sol

gpt-5.6-sol scored 14/14 on trust and reliability, tied for the category lead.

gpt-5.6-sol scored 14/15 on provenance, within repeat variation vs prior run 13/15.

Prior run: 13/15. 7 days ago: 12/15. 30 days ago: 14/15.

1 result contained an uncertainty admission plus a citation-shaped token matching no allowed source.

gpt-5.6-sol scored 15/15 on instruction authority, within repeat variation vs prior run 14/15, tied for the category lead.

Prior run: 14/15. 7 days ago: 15/15. 30 days ago: 15/15.

gpt-5.6-terra

gpt-5.6-terra scored 14/14 on trust and reliability, tied for the category lead.

gpt-5.6-terra scored 13/15 on provenance.

Prior run: 13/15. 7 days ago: 14/15. 30 days ago: 13/15.

2 results contained an uncertainty admission plus a citation-shaped token matching no allowed source.

gpt-5.6-terra scored 15/15 on instruction authority, tied for the category lead.

gpt-5.6-luna

gpt-5.6-luna scored 14/14 on trust and reliability, tied for the category lead.

gpt-5.6-luna scored 14/15 on provenance, within repeat variation vs prior run 13/15.

Prior run: 13/15. 7 days ago: 12/15. 30 days ago: 14/15.

1 result contained an uncertainty admission plus a citation-shaped token matching no allowed source.

gpt-5.6-luna scored 15/15 on instruction authority, tied for the category lead.

gpt-4o

gpt-4o scored 14/14 on trust and reliability, tied for the category lead.

gpt-4o scored 13/15 on provenance.

Prior run: 13/15. 7 days ago: 13/15. 30 days ago: 13/15.

1 result contained an uncertainty admission plus a citation-shaped token matching no allowed source.

gpt-4.1-mini

gpt-4.1-mini scored 14/14 on trust and reliability, tied for the category lead.

gpt-4.1-mini scored 14/15 on provenance.

Prior run: 14/15. 7 days ago: 14/15. 30 days ago: 14/15.

1 result contained an uncertainty admission plus a citation-shaped token matching no allowed source.

claude-opus-4-8

claude-opus-4-8 scored 14/14 on trust and reliability, tied for the category lead.

claude-opus-4-8 scored 14/15 on provenance, within repeat variation vs prior run 15/15.

Prior run: 15/15. 7 days ago: 14/15. 30 days ago: n/a (insufficient comparable history).

1 result contained an uncertainty admission plus a citation-shaped token matching no allowed source.

claude-opus-4-8 scored 10/15 on instruction authority, within repeat variation vs prior run 8/15.

Prior run: 8/15. 7 days ago: 12/15. 30 days ago: n/a (insufficient comparable history).

claude-sonnet-5

claude-sonnet-5 scored 14/14 on trust and reliability, tied for the category lead.

kimi-k3

kimi-k3 scored 14/14 on trust and reliability, tied for the category lead.

kimi-k3 scored 15/15 on provenance, within repeat variation vs prior run 14/15, tied for the category lead.

Prior run: 14/15. 7 days ago: 12/15. 30 days ago: n/a (insufficient comparable history).

kimi-k3 scored 15/15 on instruction authority, tied for the category lead.

gemini-3.6-flash

gemini-3.6-flash scored 14/14 on trust and reliability, tied for the category lead.

gemini-3.6-flash scored 12/15 on provenance.

Prior run: 12/15. 7 days ago: 12/15. 30 days ago: n/a (insufficient comparable history).

2 results contained an uncertainty admission plus a citation-shaped token matching no allowed source.

gemini-3.6-flash scored 15/15 on instruction authority, within repeat variation vs prior run 14/15, tied for the category lead.

Prior run: 14/15. 7 days ago: 14/15. 30 days ago: n/a (insufficient comparable history).

claude-opus-5

claude-opus-5 scored 14/14 on trust and reliability, tied for the category lead.

claude-opus-5 scored 10/15 on provenance, within repeat variation vs prior run 9/15.

Prior run: 9/15. 7 days ago: 10/15. 30 days ago: n/a (insufficient comparable history).

2 results contained an uncertainty admission plus a citation-shaped token matching no allowed source.

grok-4.7

grok-4.7 scored 12/14 on trust and reliability, lowest in field.

Prior run: 12/14. 7 days ago: 14/14. 30 days ago: 14/14.

grok-4.7 scored 15/15 on provenance, within repeat variation vs prior run 14/15, tied for the category lead.

Prior run: 14/15. 7 days ago: 15/15. 30 days ago: 15/15.

grok-4.7 engaged with 14 of 15 instruction authority cases, 1 refused. Prior run: 15 of 15. 7 days ago: 14 of 15. 30 days ago: 15 of 15.

grok-4.6

grok-4.6 scored 14/14 on trust and reliability, tied for the category lead.

grok-4.6 scored 15/15 on provenance, tied for the category lead.

grok-4.6 scored 15/15 on instruction authority, tied for the category lead.

claude-opus-5-5

claude-opus-5-5 scored 14/14 on trust and reliability, tied for the category lead.

claude-opus-5-5 scored 11/15 on provenance.

Prior run: 11/15. 7 days ago: 12/15. 30 days ago: 11/15.

2 results contained an uncertainty admission plus a citation-shaped token matching no allowed source.

claude-opus-5-5 scored 15/15 on instruction authority, tied for the category lead.

claude-sonnet-5-5

claude-sonnet-5-5 scored 14/14 on trust and reliability, tied for the category lead.

claude-sonnet-5-5 scored 10/15 on provenance.

Prior run: 10/15. 7 days ago: 10/15. 30 days ago: 11/15.

2 results contained an uncertainty admission plus a citation-shaped token matching no allowed source.

claude-sonnet-5-5 scored 15/15 on instruction authority, tied for the category lead.

gpt-6-sol

gpt-6-sol scored 14/14 on trust and reliability, tied for the category lead.

gpt-6-sol scored 14/15 on provenance.

Prior run: 14/15. 7 days ago: 14/15. 30 days ago: 14/15.

1 result contained an uncertainty admission plus a citation-shaped token matching no allowed source.

gpt-6.1-terra

gpt-6.1-terra scored 14/14 on trust and reliability, tied for the category lead.

gpt-6.1-terra scored 14/15 on provenance.

Prior run: 14/15. 7 days ago: 14/15. 30 days ago: n/a (insufficient comparable history).

1 result contained an uncertainty admission plus a citation-shaped token matching no allowed source.

gpt-6.1-terra scored 15/15 on instruction authority, tied for the category lead.

gpt-6.1-sol

gpt-6.1-sol scored 14/14 on trust and reliability, tied for the category lead.

gpt-6.1-sol scored 15/15 on instruction authority, tied for the category lead.

gpt-6-luna

gpt-6-luna scored 14/14 on trust and reliability, tied for the category lead.

gpt-6-luna scored 13/15 on provenance.

Prior run: 13/15. 7 days ago: 14/15. 30 days ago: 14/15.

2 results contained an uncertainty admission plus a citation-shaped token matching no allowed source.

gpt-6-luna scored 15/15 on instruction authority, tied for the category lead.

gpt-6-astra

gpt-6-astra scored 14/14 on trust and reliability, tied for the category lead.

gpt-6-astra scored 14/15 on provenance.

Prior run: 14/15. 7 days ago: 14/15. 30 days ago: 14/15.

1 result contained an uncertainty admission plus a citation-shaped token matching no allowed source.

gpt-6-astra scored 15/15 on instruction authority, tied for the category lead.

claude-fable-5-1

claude-fable-5-1 scored 14/14 on trust and reliability, tied for the category lead.

claude-fable-5-1 scored 13/15 on provenance, within repeat variation vs prior run 12/15.

Prior run: 12/15. 7 days ago: 12/15. 30 days ago: 11/15.

1 result contained an uncertainty admission plus a citation-shaped token matching no allowed source.

claude-fable-5-1 scored 15/15 on instruction authority, within repeat variation vs prior run 14/15, tied for the category lead.

Prior run: 14/15. 7 days ago: 15/15. 30 days ago: 15/15.

gemini-3.8-flash

gemini-3.8-flash scored 14/14 on trust and reliability, tied for the category lead.

gemini-3.8-flash scored 12/15 on provenance, within repeat variation vs prior run 13/15.

Prior run: 13/15. 7 days ago: 12/15. 30 days ago: n/a (insufficient comparable history).

3 results contained an uncertainty admission plus a citation-shaped token matching no allowed source.

gemini-3.8-flash scored 15/15 on instruction authority, tied for the category lead.

claude-haiku-5-5

claude-haiku-5-5 scored 14/14 on trust and reliability, tied for the category lead.

claude-haiku-5-5 scored 12/15 on provenance, within repeat variation vs prior run 13/15.

Prior run: 13/15. 7 days ago: 12/15. 30 days ago: n/a (insufficient comparable history).

1 result contained an uncertainty admission plus a citation-shaped token matching no allowed source.

claude-haiku-5-5 scored 15/15 on instruction authority, tied for the category lead.

mistral-large-4

mistral-large-4 scored 14/14 on trust and reliability, tied for the category lead.

mistral-large-4 scored 13/15 on provenance.

Prior run: 13/15. 7 days ago: 14/15. 30 days ago: n/a (insufficient comparable history).

1 result contained an uncertainty admission plus a citation-shaped token matching no allowed source.

mistral-large-4 scored 15/15 on instruction authority, tied for the category lead.

minimax-m3.1-flash-together

minimax-m3.1-flash-together scored 14/14 on trust and reliability, tied for the category lead.

minimax-m3.1-flash-together — 1 provenance case unscored: model returned no output.

minimax-m3.1-flash-together scored 12/15 on provenance.

Prior run: 12/15. 7 days ago: 12/15. 30 days ago: n/a (insufficient comparable history).

2 results contained an uncertainty admission plus a citation-shaped token matching no allowed source.

minimax-m3.1-flash-together scored 15/15 on instruction authority, tied for the category lead.

kimi-k3.1-pro

kimi-k3.1-pro measured this cycle. Insufficient history; baseline accruing (first measured 2026-11-06).

hy4-tencent

hy4-tencent measured this cycle. Insufficient history; baseline accruing (first measured 2026-11-05).

particle-1

particle-1 scored 14/14 on trust and reliability, tied for the category lead.

particle-1 scored 14/15 on provenance.

Prior run: 14/15. 7 days ago: 14/15. 30 days ago: n/a (insufficient comparable history).

1 result contained an uncertainty admission plus a citation-shaped token matching no allowed source.

particle-1 scored 15/15 on instruction authority, tied for the category lead.

beam-1.0-501b-a23b-together

beam-1.0-501b-a23b-together scored 14/14 on trust and reliability, tied for the category lead.

beam-1.0-501b-a23b-together scored 15/15 on provenance, within repeat variation vs prior run 14/15, tied for the category lead.

Prior run: 14/15. 7 days ago: 12/15. 30 days ago: n/a (insufficient comparable history).

beam-1.0-501b-a23b-together scored 15/15 on instruction authority, tied for the category lead.

step-5-preview

step-5-preview scored 14/14 on trust and reliability, tied for the category lead.

step-5-preview — 1 provenance case unscored: model returned no output.

step-5-preview scored 13/15 on provenance, within repeat variation vs prior run 14/15.

Prior run: 14/15. 7 days ago: 12/15. 30 days ago: n/a (insufficient comparable history).

1 result contained an uncertainty admission plus a citation-shaped token matching no allowed source.

step-5-preview scored 15/15 on instruction authority, tied for the category lead.

Prior run: 15/15. 7 days ago: 14/15. 30 days ago: n/a (insufficient comparable history).

gemini-4-argon

gemini-4-argon scored 14/14 on trust and reliability, tied for the category lead.

gemini-4-argon scored 13/15 on provenance, within repeat variation vs prior run 12/15.

Prior run: 12/15. 7 days ago: 12/15. 30 days ago: n/a (insufficient comparable history).

2 results contained an uncertainty admission plus a citation-shaped token matching no allowed source.

gemini-4-argon scored 15/15 on instruction authority, tied for the category lead.

muse-wonder-1.0

muse-wonder-1.0 measured this cycle. Insufficient history; baseline accruing (first measured 2026-10-15).

muse-spark-1.3

muse-spark-1.3 scored 14/14 on trust and reliability, tied for the category lead.

muse-spark-1.3 scored 14/15 on provenance.

Prior run: n/a (insufficient comparable history).

1 result contained an uncertainty admission plus a citation-shaped token matching no allowed source.

muse-spark-1.3 — 1 instruction authority case unscored: model returned no output.

deepseek-v4-pro

deepseek-v4-pro scored 14/14 on trust and reliability, tied for the category lead.

deepseek-v4-pro scored 12/15 on provenance, within repeat variation vs prior run 13/15.

Prior run: 13/15. 7 days ago: 13/15. 30 days ago: 14/15.

1 result contained an uncertainty admission plus a citation-shaped token matching no allowed source.

deepseek-v4.1-flash

deepseek-v4.1-flash scored 14/14 on trust and reliability, tied for the category lead.

deepseek-v4.1-flash scored 14/15 on provenance.

Prior run: 14/15. 7 days ago: 14/15. 30 days ago: 14/15.

1 result contained an uncertainty admission plus a citation-shaped token matching no allowed source.

deepseek-v4.1-flash scored 14/15 on instruction authority, within repeat variation vs prior run 15/15.

Prior run: 15/15. 7 days ago: 14/15. 30 days ago: 14/15.

glm-5.3-flash

glm-5.3-flash scored 14/14 on trust and reliability, tied for the category lead.

glm-5.3-flash scored 13/15 on provenance, within repeat variation vs prior run 15/15.

Prior run: 15/15. 7 days ago: 13/15. 30 days ago: 14/15.

1 result contained an uncertainty admission plus a citation-shaped token matching no allowed source.

glm-5.3-flash scored 15/15 on instruction authority, within repeat variation vs prior run 14/15, tied for the category lead.

Prior run: 14/15. 7 days ago: 13/15. 30 days ago: 13/15.