AI Intelligence

Model performance on controlled practical tasks.

1,224 task results
3 model families
17 historical configurations

Model × reasoning effort

Recorded ability. Select a value to filter the ranking.

Recorded ability by model and reasoning effort
Effort	Luna	Sol	Terra
low	69.406	76.358	69.978
medium	70.594	77.588	74.451
high	72.665	73.219	74.602
xhigh	75.747	75.929	71.614
max	76.314	75.225	71.572
ultra	—	75.225	72.789

69.40677.588

Point estimates · — Not measured.

说明

All 17 configurations stay visible here. Color compares point estimates; overlapping intervals in the ranking do not establish a statistical winner.

Practical task results

How scoring works

Ability: calibrated task score. Strict pass: fully correct outcomes. Higher is better; lines show conditional 95% intervals.

AIQ Core 1.0.7 historical results. Rank uses calibrated ability across all 17 configurations in this dataset. Equal recorded scores share a rank.
Rank				Coverage % of tasks	Evidence
1	Sol mediumgpt-5.6-sol	77.58865.0–86.6	27.8%20/72	100.0%72/72	Result evidence Sol · medium Core 1.0.7 · Historical measurement · Rank 1 of 17 Ability 0–100 77.58865.0–86.6 · conditional 95% interval Strict pass % of scored tasks 27.8%20/72 scored tasks Gap to top score ability points 0.000 Task coverage observed / expected 72/72 Compared with Sol medium (77.588) in this dataset. Point estimates do not establish a statistical winner. Intervals & interpretation Ability interval conditional 95% 65.0–86.6 Strict pass interval 95% · % 18.8–39.0% Coverage % of tasks 100.0% The ability interval holds the item bank fixed. Historical measurements retain their original task and model versions. They do not qualify as current Core 1.1.0 results. Run identity & original JSON Model gpt-5.6-sol Recorded date 2026-08-28 Scoring version 1.0.8 Run ID `run_5a2a4eca1cb69ceb798ada0d14415156ac4ab27f458ab646760fecac602b31b1` Open original public record ↗
2	Sol lowgpt-5.6-sol	76.35863.5–85.7	26.4%19/72	100.0%72/72	Result evidence Sol · low Core 1.0.7 · Historical measurement · Rank 2 of 17 Ability 0–100 76.35863.5–85.7 · conditional 95% interval Strict pass % of scored tasks 26.4%19/72 scored tasks Gap to top score ability points -1.230 Task coverage observed / expected 72/72 Compared with Sol medium (77.588) in this dataset. Point estimates do not establish a statistical winner. Intervals & interpretation Ability interval conditional 95% 63.5–85.7 Strict pass interval 95% · % 17.6–37.6% Coverage % of tasks 100.0% The ability interval holds the item bank fixed. Historical measurements retain their original task and model versions. They do not qualify as current Core 1.1.0 results. Run identity & original JSON Model gpt-5.6-sol Recorded date 2026-08-28 Scoring version 1.0.8 Run ID `run_602248d32fb259e607d90a9fc14707df48e308f4ede3921cc6324341cde06900` Open original public record ↗
3	Luna maxgpt-5.6-luna	76.31463.4–85.7	27.8%20/72	100.0%72/72	Result evidence Luna · max Core 1.0.7 · Historical measurement · Rank 3 of 17 Ability 0–100 76.31463.4–85.7 · conditional 95% interval Strict pass % of scored tasks 27.8%20/72 scored tasks Gap to top score ability points -1.274 Task coverage observed / expected 72/72 Compared with Sol medium (77.588) in this dataset. Point estimates do not establish a statistical winner. Intervals & interpretation Ability interval conditional 95% 63.4–85.7 Strict pass interval 95% · % 18.8–39.0% Coverage % of tasks 100.0% The ability interval holds the item bank fixed. Historical measurements retain their original task and model versions. They do not qualify as current Core 1.1.0 results. Run identity & original JSON Model gpt-5.6-luna Recorded date 2026-08-28 Scoring version 1.0.8 Run ID `run_cc3cfefe0e6c96deb38f020dcb5154bda98b62333adc00bd816f9c9f6c7232ab` Open original public record ↗
4	Sol xhighgpt-5.6-sol	75.92963.0–85.4	27.8%20/72	100.0%72/72	Result evidence Sol · xhigh Core 1.0.7 · Historical measurement · Rank 4 of 17 Ability 0–100 75.92963.0–85.4 · conditional 95% interval Strict pass % of scored tasks 27.8%20/72 scored tasks Gap to top score ability points -1.659 Task coverage observed / expected 72/72 Compared with Sol medium (77.588) in this dataset. Point estimates do not establish a statistical winner. Intervals & interpretation Ability interval conditional 95% 63.0–85.4 Strict pass interval 95% · % 18.8–39.0% Coverage % of tasks 100.0% The ability interval holds the item bank fixed. Historical measurements retain their original task and model versions. They do not qualify as current Core 1.1.0 results. Run identity & original JSON Model gpt-5.6-sol Recorded date 2026-08-28 Scoring version 1.0.8 Run ID `run_b2372624b77abd0598e1c4f3b783a3726ec32241eacc3396d803f4ae0641b430` Open original public record ↗
5	Luna xhighgpt-5.6-luna	75.74762.8–85.3	29.2%21/72	100.0%72/72	Result evidence Luna · xhigh Core 1.0.7 · Historical measurement · Rank 5 of 17 Ability 0–100 75.74762.8–85.3 · conditional 95% interval Strict pass % of scored tasks 29.2%21/72 scored tasks Gap to top score ability points -1.841 Task coverage observed / expected 72/72 Compared with Sol medium (77.588) in this dataset. Point estimates do not establish a statistical winner. Intervals & interpretation Ability interval conditional 95% 62.8–85.3 Strict pass interval 95% · % 19.9–40.5% Coverage % of tasks 100.0% The ability interval holds the item bank fixed. Historical measurements retain their original task and model versions. They do not qualify as current Core 1.1.0 results. Run identity & original JSON Model gpt-5.6-luna Recorded date 2026-08-28 Scoring version 1.0.8 Run ID `run_652a8005e9e3066cf1987e27a140a421e256f7961aca372489ff0686f885ed83` Open original public record ↗
6	Sol maxgpt-5.6-sol	75.22562.2–84.9	26.4%19/72	100.0%72/72	Result evidence Sol · max Core 1.0.7 · Historical measurement · Rank 6 of 17 Ability 0–100 75.22562.2–84.9 · conditional 95% interval Strict pass % of scored tasks 26.4%19/72 scored tasks Gap to top score ability points -2.363 Task coverage observed / expected 72/72 Compared with Sol medium (77.588) in this dataset. Point estimates do not establish a statistical winner. Intervals & interpretation Ability interval conditional 95% 62.2–84.9 Strict pass interval 95% · % 17.6–37.6% Coverage % of tasks 100.0% The ability interval holds the item bank fixed. Historical measurements retain their original task and model versions. They do not qualify as current Core 1.1.0 results. Run identity & original JSON Model gpt-5.6-sol Recorded date 2026-08-28 Scoring version 1.0.8 Run ID `run_366299d303a738a36a1289a57a3234cc7059d0b057e620f6139ac02787899be3` Open original public record ↗
6	Sol ultragpt-5.6-sol	75.22562.2–84.9	27.8%20/72	100.0%72/72	Result evidence Sol · ultra Core 1.0.7 · Historical measurement · Rank 6 of 17 Ability 0–100 75.22562.2–84.9 · conditional 95% interval Strict pass % of scored tasks 27.8%20/72 scored tasks Gap to top score ability points -2.363 Task coverage observed / expected 72/72 Compared with Sol medium (77.588) in this dataset. Point estimates do not establish a statistical winner. Intervals & interpretation Ability interval conditional 95% 62.2–84.9 Strict pass interval 95% · % 18.8–39.0% Coverage % of tasks 100.0% The ability interval holds the item bank fixed. Historical measurements retain their original task and model versions. They do not qualify as current Core 1.1.0 results. Run identity & original JSON Model gpt-5.6-sol Recorded date 2026-08-28 Scoring version 1.0.8 Run ID `run_395ab5e170b1b9653667e306f6c7ffd93d118175c19adacb0a7c5d5b8bfca7b1` Open original public record ↗
8	Terra highgpt-5.6-terra	74.60261.4–84.4	23.6%17/72	100.0%72/72	Result evidence Terra · high Core 1.0.7 · Historical measurement · Rank 8 of 17 Ability 0–100 74.60261.4–84.4 · conditional 95% interval Strict pass % of scored tasks 23.6%17/72 scored tasks Gap to top score ability points -2.986 Task coverage observed / expected 72/72 Compared with Sol medium (77.588) in this dataset. Point estimates do not establish a statistical winner. Intervals & interpretation Ability interval conditional 95% 61.4–84.4 Strict pass interval 95% · % 15.3–34.6% Coverage % of tasks 100.0% The ability interval holds the item bank fixed. Historical measurements retain their original task and model versions. They do not qualify as current Core 1.1.0 results. Run identity & original JSON Model gpt-5.6-terra Recorded date 2026-08-28 Scoring version 1.0.8 Run ID `run_bde48d35c4ab92dc81c7d7b05f60687b45850990211853d6674a97b37d9c56fe` Open original public record ↗
9	Terra mediumgpt-5.6-terra	74.45161.3–84.3	27.8%20/72	100.0%72/72	Result evidence Terra · medium Core 1.0.7 · Historical measurement · Rank 9 of 17 Ability 0–100 74.45161.3–84.3 · conditional 95% interval Strict pass % of scored tasks 27.8%20/72 scored tasks Gap to top score ability points -3.137 Task coverage observed / expected 72/72 Compared with Sol medium (77.588) in this dataset. Point estimates do not establish a statistical winner. Intervals & interpretation Ability interval conditional 95% 61.3–84.3 Strict pass interval 95% · % 18.8–39.0% Coverage % of tasks 100.0% The ability interval holds the item bank fixed. Historical measurements retain their original task and model versions. They do not qualify as current Core 1.1.0 results. Run identity & original JSON Model gpt-5.6-terra Recorded date 2026-08-28 Scoring version 1.0.8 Run ID `run_8832ca59be523e36b05b5516ff7281b3f711eaf10cfe3581e45b8b95d1c4992b` Open original public record ↗
10	Sol highgpt-5.6-sol	73.21959.9–83.4	25.0%18/72	100.0%72/72	Result evidence Sol · high Core 1.0.7 · Historical measurement · Rank 10 of 17 Ability 0–100 73.21959.9–83.4 · conditional 95% interval Strict pass % of scored tasks 25.0%18/72 scored tasks Gap to top score ability points -4.369 Task coverage observed / expected 72/72 Compared with Sol medium (77.588) in this dataset. Point estimates do not establish a statistical winner. Intervals & interpretation Ability interval conditional 95% 59.9–83.4 Strict pass interval 95% · % 16.4–36.1% Coverage % of tasks 100.0% The ability interval holds the item bank fixed. Historical measurements retain their original task and model versions. They do not qualify as current Core 1.1.0 results. Run identity & original JSON Model gpt-5.6-sol Recorded date 2026-08-28 Scoring version 1.0.8 Run ID `run_437b45de053bfca2ff1f69ad7dc57c006818f1ab4203786cc7db5cb6cde076ac` Open original public record ↗
11	Terra ultragpt-5.6-terra	72.78959.4–83.0	25.0%18/72	100.0%72/72	Result evidence Terra · ultra Core 1.0.7 · Historical measurement · Rank 11 of 17 Ability 0–100 72.78959.4–83.0 · conditional 95% interval Strict pass % of scored tasks 25.0%18/72 scored tasks Gap to top score ability points -4.799 Task coverage observed / expected 72/72 Compared with Sol medium (77.588) in this dataset. Point estimates do not establish a statistical winner. Intervals & interpretation Ability interval conditional 95% 59.4–83.0 Strict pass interval 95% · % 16.4–36.1% Coverage % of tasks 100.0% The ability interval holds the item bank fixed. Historical measurements retain their original task and model versions. They do not qualify as current Core 1.1.0 results. Run identity & original JSON Model gpt-5.6-terra Recorded date 2026-08-28 Scoring version 1.0.8 Run ID `run_8dc4463121678365f69e79c11278f844c6b2acdd49f76da099a46e852be079ef` Open original public record ↗
12	Luna highgpt-5.6-luna	72.66559.2–82.9	27.8%20/72	100.0%72/72	Result evidence Luna · high Core 1.0.7 · Historical measurement · Rank 12 of 17 Ability 0–100 72.66559.2–82.9 · conditional 95% interval Strict pass % of scored tasks 27.8%20/72 scored tasks Gap to top score ability points -4.923 Task coverage observed / expected 72/72 Compared with Sol medium (77.588) in this dataset. Point estimates do not establish a statistical winner. Intervals & interpretation Ability interval conditional 95% 59.2–82.9 Strict pass interval 95% · % 18.8–39.0% Coverage % of tasks 100.0% The ability interval holds the item bank fixed. Historical measurements retain their original task and model versions. They do not qualify as current Core 1.1.0 results. Run identity & original JSON Model gpt-5.6-luna Recorded date 2026-08-28 Scoring version 1.0.8 Run ID `run_9c6cefcef6a726dc2ab54d6b851f06629e45097f8de1b6d47f709ff1cccee3b4` Open original public record ↗
13	Terra xhighgpt-5.6-terra	71.61458.0–82.1	26.4%19/72	100.0%72/72	Result evidence Terra · xhigh Core 1.0.7 · Historical measurement · Rank 13 of 17 Ability 0–100 71.61458.0–82.1 · conditional 95% interval Strict pass % of scored tasks 26.4%19/72 scored tasks Gap to top score ability points -5.974 Task coverage observed / expected 72/72 Compared with Sol medium (77.588) in this dataset. Point estimates do not establish a statistical winner. Intervals & interpretation Ability interval conditional 95% 58.0–82.1 Strict pass interval 95% · % 17.6–37.6% Coverage % of tasks 100.0% The ability interval holds the item bank fixed. Historical measurements retain their original task and model versions. They do not qualify as current Core 1.1.0 results. Run identity & original JSON Model gpt-5.6-terra Recorded date 2026-08-28 Scoring version 1.0.8 Run ID `run_81e97c23f7c4ad230af28152a1c110db8164e0a8769f2dc78ceadb049cf0d294` Open original public record ↗
14	Terra maxgpt-5.6-terra	71.57258.0–82.1	26.4%19/72	100.0%72/72	Result evidence Terra · max Core 1.0.7 · Historical measurement · Rank 14 of 17 Ability 0–100 71.57258.0–82.1 · conditional 95% interval Strict pass % of scored tasks 26.4%19/72 scored tasks Gap to top score ability points -6.016 Task coverage observed / expected 72/72 Compared with Sol medium (77.588) in this dataset. Point estimates do not establish a statistical winner. Intervals & interpretation Ability interval conditional 95% 58.0–82.1 Strict pass interval 95% · % 17.6–37.6% Coverage % of tasks 100.0% The ability interval holds the item bank fixed. Historical measurements retain their original task and model versions. They do not qualify as current Core 1.1.0 results. Run identity & original JSON Model gpt-5.6-terra Recorded date 2026-08-28 Scoring version 1.0.8 Run ID `run_29b8598df64fabb999b6f034956e2a10f13758a151d32272cb8cf8746b2d0c52` Open original public record ↗
15	Luna mediumgpt-5.6-luna	70.59456.9–81.4	23.6%17/72	100.0%72/72	Result evidence Luna · medium Core 1.0.7 · Historical measurement · Rank 15 of 17 Ability 0–100 70.59456.9–81.4 · conditional 95% interval Strict pass % of scored tasks 23.6%17/72 scored tasks Gap to top score ability points -6.994 Task coverage observed / expected 72/72 Compared with Sol medium (77.588) in this dataset. Point estimates do not establish a statistical winner. Intervals & interpretation Ability interval conditional 95% 56.9–81.4 Strict pass interval 95% · % 15.3–34.6% Coverage % of tasks 100.0% The ability interval holds the item bank fixed. Historical measurements retain their original task and model versions. They do not qualify as current Core 1.1.0 results. Run identity & original JSON Model gpt-5.6-luna Recorded date 2026-08-28 Scoring version 1.0.8 Run ID `run_92a9f7939a2dc7b961a3cbd6b13f719907ae2adfbaca013b6f086514266747cc` Open original public record ↗
16	Terra lowgpt-5.6-terra	69.97856.2–80.9	29.2%21/72	100.0%72/72	Result evidence Terra · low Core 1.0.7 · Historical measurement · Rank 16 of 17 Ability 0–100 69.97856.2–80.9 · conditional 95% interval Strict pass % of scored tasks 29.2%21/72 scored tasks Gap to top score ability points -7.610 Task coverage observed / expected 72/72 Compared with Sol medium (77.588) in this dataset. Point estimates do not establish a statistical winner. Intervals & interpretation Ability interval conditional 95% 56.2–80.9 Strict pass interval 95% · % 19.9–40.5% Coverage % of tasks 100.0% The ability interval holds the item bank fixed. Historical measurements retain their original task and model versions. They do not qualify as current Core 1.1.0 results. Run identity & original JSON Model gpt-5.6-terra Recorded date 2026-08-28 Scoring version 1.0.8 Run ID `run_0333c6971437abbd8e524383310d57e9441e80de282d6cd0ba85a4183602d1be` Open original public record ↗
17	Luna lowgpt-5.6-luna	69.40655.6–80.4	25.0%18/72	100.0%72/72	Result evidence Luna · low Core 1.0.7 · Historical measurement · Rank 17 of 17 Ability 0–100 69.40655.6–80.4 · conditional 95% interval Strict pass % of scored tasks 25.0%18/72 scored tasks Gap to top score ability points -8.182 Task coverage observed / expected 72/72 Compared with Sol medium (77.588) in this dataset. Point estimates do not establish a statistical winner. Intervals & interpretation Ability interval conditional 95% 55.6–80.4 Strict pass interval 95% · % 16.4–36.1% Coverage % of tasks 100.0% The ability interval holds the item bank fixed. Historical measurements retain their original task and model versions. They do not qualify as current Core 1.1.0 results. Run identity & original JSON Model gpt-5.6-luna Recorded date 2026-08-28 Scoring version 1.0.8 Run ID `run_bdc18132086ad72aa7b1f858785bf0673abb4507c9f6633eaad13d06e1b84ad0` Open original public record ↗

Point-estimate order · intervals may overlap.

说明

Ability rank stays relative to all 17 configurations when filters change. Equal recorded ability shares a rank; point-estimate order does not establish a statistical winner.

Lines show conditional 95% intervals on the shared ability scale 54.062–88.158, including after filtering. Ability and intervals use the selected benchmark version and scoring 1.0.8. Historical measurements retain their original model and task version. Ability is specific to Core tasks and is not a human IQ norm.

Full scoring method and limitations