Home › Videos › GPT-6.1 Sol vs Sonnet 5.5 — My Real App Tests

29 Sept 2026 · 2 runs · 2 models

GPT-6.1 Sol vs Sonnet 5.5 — My Real App Tests

Two AI models, six interactive web-page tests. Every score, time, token count and cost, with the pages each model built.

Watch on YouTube Compare models from this video Every score (CSV)

The results

The same scores as the leaderboard: a test that failed is left out of its run's score, not counted as zero. Time, tokens and cost cover each whole run, wildcards included.

See the whole board

Ranking

Quality v1 over measured non-wildcard pages. Different workloads: browse the outputs and coverage before comparing models.

claude-sonnet-5-5effort: defaultAnthropic · 4/5 scored pages

7.3

gpt-6.1-soleffort: defaultOpenAI · 5/5 scored pages

6.3

0510

Score against output tokens

Quality v1 against full-run tokens. Different workloads and hardware affect this comparison.

Claude SonnetGPT Sol

Output tokens include the model's thinking. Like the time, tokens and cost table, it counts every test the model took, wildcards included. A hollow dot is not an exact count: some tests reported none, or the count was worked out from the length of the output. The score axis is zoomed to the data.

Score against time

Quality v1 against full-run time. Different workloads and hardware affect this comparison.

Claude SonnetGPT Sol

Time is how long the model took to answer, added up over its tests; judging is not counted. Like the time, tokens and cost table, it counts every test the model took, wildcards included. A hollow dot is a lower bound: some tests have no time recorded. The score axis is zoomed to the data.

Where each model is strong and weak

One row per model, one column per test and per scoring criterion.

	By test					By what the judges scored
Model	Aero Lab	Jet Canyon	Blast Island	Launch Day	Alex Optics	Working behaviour	Visual quality	Engineering quality	Usability / accessibility
claude-sonnet-5-5effort: default	5.9	8.0	7.7	failed	7.6	7.9	7.6	7.1	6.0
gpt-6.1-soleffort: default	4.1	7.1	6.5	6.8	6.9	6.6	6.9	5.8	5.2

Violet is strong, orange is weak, on the same 0–10 scale in every cell.Different prompts and protocols: no best-in-column verdict.

Time, tokens and cost

What each model used across all its tests, wildcards included.

Model	Tests	Time	Output tokens	Speed	Cost	Subscription used
claude-sonnet-5-5effort: defaultAnthropic	61 failed	≥1h 43m	≥672k	132.5 tok/s	—	—
gpt-6.1-soleffort: defaultOpenAI	6	31m 42s	62k	32.9 tok/s	—	—

≈ An estimate: the provider reported no count, so it is worked out from the length of the output (about 4 characters a token). — Not measured. Cost says how it was worked out: charged by the provider, at API prices (for a subscription), or no API charge on local hardware. ≥ means some tests had no price.

Read from the account's usage window in whole percentage points, so a short run can show 0 points. It counts everything the account did in that time, not only this run.

Runs in this video

Every run behind the numbers. Scores name their comparison basis.

Model

Score / basis

Time

Output tokens

Cost

claude-sonnet-5-5effort: defaultpartial quality coverageAnthropic · 29 Sept 2026

7.3 / 10Quality v1 · 4/5 pages

Time≥1h 43m

Output tokens≥672k

Cost—

Open Compare

gpt-6.1-soleffort: defaultOpenAI · 29 Sept 2026

6.3 / 10Quality v1 · 5/5 pages

Time31m 42s

Output tokens62k

Cost—

Open Compare

claude-sonnet-5-5effort: default

Anthropic · 7.3 / 10 · Quality v1 · 4/5 pages · 1h 43m · — partial quality coverage

metrics.csv run.json Open the run page Compare

Screenshot of the page claude-sonnet-5-5 made for Aero Lab

5.9

Aero Lab claude-sonnet-5-5 · effort: default 19m 07s · 147k tokens · — Quality v1 · Working behaviour 6.5 · Visual quality 5.0 · Engineering quality 7.0 · Usability / accessibility 5.0scored by claude-code/claude-opus-5-5 (quality-1 rubric)prompt 6c713ca03954f51e

Earlier evaluations

4.4/10 · legacy · claude-code/claude-opus-5-5

Different rubrics; these values are historical, not a score change.

Open the page Details

Screenshot of the page claude-sonnet-5-5 made for Jet Canyon

8.0

Jet Canyon claude-sonnet-5-5 · effort: default 19m 43s · 162k tokens · — Quality v1 · Working behaviour 8.5 · Visual quality 9.0 · Engineering quality 7.0 · Usability / accessibility 6.5scored by claude-code/claude-opus-5-5 (quality-1 rubric)prompt 021db4866371ec15

Earlier evaluations

7.7/10 · legacy · claude-code/claude-opus-5-5

Different rubrics; these values are historical, not a score change.

Open the page Details

Screenshot of the page claude-sonnet-5-5 made for Blast Island

7.7

Blast Island claude-sonnet-5-5 · effort: default 19m 17s · 155k tokens · — Quality v1 · Working behaviour 8.5 · Visual quality 8.5 · Engineering quality 7.0 · Usability / accessibility 6.0scored by claude-code/claude-opus-5-5 (quality-1 rubric)prompt a282381696456fc9

Earlier evaluations

7.7/10 · legacy · claude-code/claude-opus-5-5

Different rubrics; these values are historical, not a score change.

Open the page Details

No screenshot keptfailed

Launch Daybest of 2 claude-sonnet-5-5 · effort: default 18m 37s · — tokens · — Legacy rubric · Functional 0.0 · Visual 0.0 · Code 0.0 · UX 0.0 · Innovation 0.0prompt 2787714823d7edc7 Details

Screenshot of the page claude-sonnet-5-5 made for Alex Optics

7.6

Alex Optics claude-sonnet-5-5 · effort: default 18m 00s · 142k tokens · — Quality v1 · Working behaviour 8.0 · Visual quality 8.0 · Engineering quality 7.5 · Usability / accessibility 6.5scored by claude-code/claude-opus-5-5 (quality-1 rubric)prompt e519ec9dadc0a186

Earlier evaluations

6.9/10 · legacy · claude-code/claude-opus-5-5

Different rubrics; these values are historical, not a score change.

Open the page Details

Screenshot of the page claude-sonnet-5-5 made for Human Pyramid Balance Puzzle

7.0

Human Pyramid Balance Puzzlewildcard claude-sonnet-5-5 · effort: default 8m 20s · 66k tokens · — Quality v1 · Working behaviour 8.0 · Visual quality 7.5 · Engineering quality 6.5 · Usability / accessibility 5.0scored by claude-code/claude-opus-5-5 (quality-1 rubric)prompt b2b5473f9f99c565

Earlier evaluations

7.7/10 · legacy · claude-code/claude-opus-5-5

Different rubrics; these values are historical, not a score change.

Open the page Details

gpt-6.1-soleffort: default

OpenAI · 6.3 / 10 · Quality v1 · 5/5 pages · 31m 42s · —

metrics.csv run.json Open the run page Compare

Screenshot of the page gpt-6.1-sol made for Aero Lab

4.1

Aero Lab gpt-6.1-sol · effort: default 4m 08s · 8.2k tokens · — Quality v1 · Working behaviour 3.5 · Visual quality 4.0 · Engineering quality 5.0 · Usability / accessibility 4.0scored by claude-code/claude-opus-5-5 (quality-1 rubric)prompt 6c713ca03954f51e

Earlier evaluations

5.8/10 · legacy · codex-cli/gpt-6.1-sol

Different rubrics; these values are historical, not a score change.

Open the page Details

Screenshot of the page gpt-6.1-sol made for Jet Canyon

7.1

Jet Canyon gpt-6.1-sol · effort: default 4m 23s · 8.6k tokens · — Quality v1 · Working behaviour 8.0 · Visual quality 8.0 · Engineering quality 6.0 · Usability / accessibility 5.5scored by claude-code/claude-opus-5-5 (quality-1 rubric)prompt 021db4866371ec15

Earlier evaluations

5.8/10 · legacy · codex-cli/gpt-6.1-sol

Different rubrics; these values are historical, not a score change.

Open the page Details

Screenshot of the page gpt-6.1-sol made for Blast Island

6.5

Blast Island gpt-6.1-sol · effort: default 5m 24s · 11k tokens · — Quality v1 · Working behaviour 7.0 · Visual quality 7.5 · Engineering quality 5.5 · Usability / accessibility 5.0scored by claude-code/claude-opus-5-5 (quality-1 rubric)prompt a282381696456fc9

Earlier evaluations

6.1/10 · legacy · codex-cli/gpt-6.1-sol

Different rubrics; these values are historical, not a score change.

Open the page Details

Screenshot of the page gpt-6.1-sol made for Launch Day

6.8

Launch Day gpt-6.1-sol · effort: default 5m 48s · 11k tokens · — Quality v1 · Working behaviour 7.5 · Visual quality 7.5 · Engineering quality 6.0 · Usability / accessibility 5.5scored by claude-code/claude-opus-5-5 (quality-1 rubric)prompt 2787714823d7edc7

Earlier evaluations

5.4/10 · legacy · codex-cli/gpt-6.1-sol

Different rubrics; these values are historical, not a score change.

Open the page Details

Screenshot of the page gpt-6.1-sol made for Alex Optics

6.9

Alex Optics gpt-6.1-sol · effort: default 8m 42s · 17k tokens · — Quality v1 · Working behaviour 7.0 · Visual quality 7.5 · Engineering quality 6.5 · Usability / accessibility 6.0scored by claude-code/claude-opus-5-5 (quality-1 rubric)prompt e519ec9dadc0a186

Earlier evaluations

5.7/10 · legacy · codex-cli/gpt-6.1-sol

Different rubrics; these values are historical, not a score change.

Open the page Details

Screenshot of the page gpt-6.1-sol made for Wind Loom Portraits

5.9

Wind Loom Portraitswildcard gpt-6.1-sol · effort: default 3m 16s · 6.4k tokens · — Quality v1 · Working behaviour 5.5 · Visual quality 7.0 · Engineering quality 5.5 · Usability / accessibility 5.0scored by claude-code/claude-opus-5-5 (quality-1 rubric)prompt 408e6b279c62918a

Earlier evaluations

7.8/10 · legacy · codex-cli/gpt-6.1-sol

Different rubrics; these values are historical, not a score change.

Open the page Details