ModelVerdict

Comparison

Gemini 3.8 Flash vs GPT-6 Sol

Same documents, same prompt: where each model gets it right, what it really costs, and how fast it responds.

Verdict

Gemini 3.8 Flash is more accurate and cheaper in 12 of 30 combinations.

More accurate

Gemini 3.8 Flash

wins 10 of 30

Cheaper

Gemini 3.8 Flash

cheaper in 12 of 30

Faster

GPT-6 Sol

2.9 sec vs 3.2 sec

Fewer fixes

Gemini 3.8 Flash

7 vs 13 documents per 1,000

Across 30 task and document-language combinations (speed and fixes: median).

At a glance

English (US)
Accuracy (correct fields)Per 1,000 documents
Task · documentsGemini 3.8 FlashGPT-6 SolDifferenceGemini 3.8 FlashGPT-6 SolDifference
Invoice extraction99.82%99.94%▲ 0.12$3.84$9.65+152 %
Email triage98.74%99.04%▲ 0.3$2.39$2.44+2 %
Personal data detection99.71%99.71%±0$1.86$2.09+12 %
Contract clauses99.90%100.00%▲ 0.1$2.47$3.33+35 %
Document summaries100.00%99.56%▼ 0.44$5.46$3.85−30 %
Questions about a document99.89%100.00%▲ 0.11$3.00$2.36−21 %
Czech
Accuracy (correct fields)Per 1,000 documents
Task · documentsGemini 3.8 FlashGPT-6 SolDifferenceGemini 3.8 FlashGPT-6 SolDifference
Invoice extraction99.95%99.90%▼ 0.05$4.26$10.12+138 %
Email triage99.19%98.96%▼ 0.23$3.06$3.19+4 %
Personal data detection99.52%99.71%▲ 0.19$2.19$2.22+2 %
Contract clauses100.00%100.00%±0$3.09$4.62+50 %
Document summaries100.00%100.00%±0$6.69$4.72−29 %
Questions about a document100.00%100.00%±0$4.36$3.95−10 %
German
Accuracy (correct fields)Per 1,000 documents
Task · documentsGemini 3.8 FlashGPT-6 SolDifferenceGemini 3.8 FlashGPT-6 SolDifference
Invoice extraction97.23%99.95%▲ 2.72$7.15$10.65+49 %
Email triage98.37%98.22%▼ 0.15$2.74$2.73−1 %
Personal data detection99.71%99.71%±0$2.12$2.13±0 %
Contract clauses99.90%100.00%▲ 0.1$3.42$4.36+28 %
Document summaries100.00%100.00%±0$6.11$4.11−33 %
Questions about a document100.00%100.00%±0$3.76$3.23−14 %
Spanish
Accuracy (correct fields)Per 1,000 documents
Task · documentsGemini 3.8 FlashGPT-6 SolDifferenceGemini 3.8 FlashGPT-6 SolDifference
Invoice extraction97.28%99.23%▲ 1.95$7.84$10.79+38 %
Email triage98.96%98.44%▼ 0.52$3.30$2.77−16 %
Personal data detection99.81%99.81%±0$2.11$2.21+5 %
Contract clauses100.00%100.00%±0$3.33$4.48+35 %
Document summaries100.00%97.89%▼ 2.11$6.78$5.02−26 %
Questions about a document100.00%100.00%±0$4.05$3.36−17 %
French
Accuracy (correct fields)Per 1,000 documents
Task · documentsGemini 3.8 FlashGPT-6 SolDifferenceGemini 3.8 FlashGPT-6 SolDifference
Invoice extraction99.95%99.90%▼ 0.05$6.07$11.00+81 %
Email triage98.59%98.59%±0$2.63$2.81+7 %
Personal data detection99.81%99.71%▼ 0.1$2.13$2.23+4 %
Contract clauses100.00%99.59%▼ 0.41$3.50$4.49+28 %
Document summaries100.00%97.22%▼ 2.78$7.06$4.58−35 %
Questions about a document100.00%100.00%±0$4.06$3.43−16 %

Difference: GPT-6 Sol vs. Gemini 3.8 Flash; accuracy in percentage points, price in percent. The better value is in bold. Click a task to see the details.

Invoice extraction · English (US) (150 documents)

GPT-6 Sol is more accurate by 0.1 pp. Gemini 3.8 Flash is 2.5× cheaper.

MetricGemini 3.8 FlashGPT-6 Sol
Correct fields99.82%99.94%
Error-free documents98.00%99.33%
Per 1,000 documents$3.84$9.65
To fix / 1,000207
Speed (median)3.2 sec4.1 sec
Valid answers100.00%100.00%

Free with login

Accuracy field by field and the documents where the models differ.

Just your email, no password.

Invoice extraction · Czech (150 documents)

Gemini 3.8 Flash is more accurate by 0.1 pp. Gemini 3.8 Flash is 2.4× cheaper.

MetricGemini 3.8 FlashGPT-6 Sol
Correct fields99.95%99.90%
Error-free documents99.33%98.67%
Per 1,000 documents$4.26$10.12
To fix / 1,000713
Speed (median)3.5 sec3.9 sec
Valid answers100.00%100.00%

Free with login

Accuracy field by field and the documents where the models differ.

Just your email, no password.

Invoice extraction · German (150 documents)

GPT-6 Sol is more accurate by 2.7 pp. Gemini 3.8 Flash is 1.5× cheaper.

MetricGemini 3.8 FlashGPT-6 Sol
Correct fields97.23%99.95%
Error-free documents96.00%99.33%
Per 1,000 documents$7.15$10.65
To fix / 1,000407
Speed (median)4.7 sec4.9 sec
Valid answers97.33%100.00%

Free with login

Accuracy field by field and the documents where the models differ.

Just your email, no password.

Invoice extraction · Spanish (150 documents)

GPT-6 Sol is more accurate by 1.9 pp. Gemini 3.8 Flash is 1.4× cheaper.

MetricGemini 3.8 FlashGPT-6 Sol
Correct fields97.28%99.23%
Error-free documents96.67%90.00%
Per 1,000 documents$7.84$10.79
To fix / 1,00033100
Speed (median)5.7 sec5.9 sec
Valid answers97.33%100.00%

Free with login

Accuracy field by field and the documents where the models differ.

Just your email, no password.

Invoice extraction · French (150 documents)

Gemini 3.8 Flash is more accurate by 0.1 pp. Gemini 3.8 Flash is 1.8× cheaper.

MetricGemini 3.8 FlashGPT-6 Sol
Correct fields99.95%99.90%
Error-free documents99.33%98.67%
Per 1,000 documents$6.07$11.00
To fix / 1,000713
Speed (median)5.3 sec6.0 sec
Valid answers100.00%100.00%

Free with login

Accuracy field by field and the documents where the models differ.

Just your email, no password.

Email triage · English (US) (150 documents)

GPT-6 Sol is more accurate by 0.3 pp.

MetricGemini 3.8 FlashGPT-6 Sol
Correct fields98.74%99.04%
Error-free documents88.67%92.00%
Per 1,000 documents$2.39$2.44
To fix / 1,00011380
Speed (median)2.2 sec2.8 sec
Valid answers100.00%100.00%

Free with login

Accuracy field by field and the documents where the models differ.

Just your email, no password.

Email triage · Czech (150 documents)

Gemini 3.8 Flash is more accurate by 0.2 pp.

MetricGemini 3.8 FlashGPT-6 Sol
Correct fields99.19%98.96%
Error-free documents92.67%90.67%
Per 1,000 documents$3.06$3.19
To fix / 1,0007393
Speed (median)2.6 sec2.8 sec
Valid answers100.00%100.00%

Free with login

Accuracy field by field and the documents where the models differ.

Just your email, no password.

Email triage · German (150 documents)

Gemini 3.8 Flash is more accurate by 0.2 pp.

MetricGemini 3.8 FlashGPT-6 Sol
Correct fields98.37%98.22%
Error-free documents85.33%84.00%
Per 1,000 documents$2.74$2.73
To fix / 1,000147160
Speed (median)2.9 sec2.9 sec
Valid answers100.00%100.00%

Free with login

Accuracy field by field and the documents where the models differ.

Just your email, no password.

Email triage · Spanish (150 documents)

Gemini 3.8 Flash is more accurate by 0.5 pp. GPT-6 Sol is 1.2× cheaper.

MetricGemini 3.8 FlashGPT-6 Sol
Correct fields98.96%98.44%
Error-free documents90.67%86.00%
Per 1,000 documents$3.30$2.77
To fix / 1,00093140
Speed (median)3.3 sec3.3 sec
Valid answers100.00%100.00%

Free with login

Accuracy field by field and the documents where the models differ.

Just your email, no password.

Email triage · French (150 documents)

Both are equally accurate. Gemini 3.8 Flash is 1.1× cheaper.

MetricGemini 3.8 FlashGPT-6 Sol
Correct fields98.59%98.59%
Error-free documents87.33%87.33%
Per 1,000 documents$2.63$2.81
To fix / 1,000127127
Speed (median)2.7 sec3.2 sec
Valid answers100.00%100.00%

Free with login

Accuracy field by field and the documents where the models differ.

Just your email, no password.

Personal data detection · English (US) (150 documents)

Both are equally accurate. Gemini 3.8 Flash is 1.1× cheaper.

MetricGemini 3.8 FlashGPT-6 Sol
Correct fields99.71%99.71%
Error-free documents98.00%98.00%
Per 1,000 documents$1.86$2.09
To fix / 1,0002020
Speed (median)1.9 sec2.6 sec
Valid answers100.00%100.00%

Free with login

Accuracy field by field and the documents where the models differ.

Just your email, no password.

Personal data detection · Czech (150 documents)

GPT-6 Sol is more accurate by 0.2 pp.

MetricGemini 3.8 FlashGPT-6 Sol
Correct fields99.52%99.71%
Error-free documents96.67%98.00%
Per 1,000 documents$2.19$2.22
To fix / 1,0003320
Speed (median)2.0 sec2.5 sec
Valid answers100.00%100.00%

Free with login

Accuracy field by field and the documents where the models differ.

Just your email, no password.

Personal data detection · German (150 documents)

Both are equally accurate.

MetricGemini 3.8 FlashGPT-6 Sol
Correct fields99.71%99.71%
Error-free documents98.00%98.00%
Per 1,000 documents$2.12$2.13
To fix / 1,0002020
Speed (median)2.3 sec2.6 sec
Valid answers100.00%100.00%

Free with login

Accuracy field by field and the documents where the models differ.

Just your email, no password.

Personal data detection · Spanish (150 documents)

Both are equally accurate.

MetricGemini 3.8 FlashGPT-6 Sol
Correct fields99.81%99.81%
Error-free documents98.67%98.67%
Per 1,000 documents$2.11$2.21
To fix / 1,0001313
Speed (median)2.3 sec2.9 sec
Valid answers100.00%100.00%

Free with login

Accuracy field by field and the documents where the models differ.

Just your email, no password.

Personal data detection · French (150 documents)

Gemini 3.8 Flash is more accurate by 0.1 pp.

MetricGemini 3.8 FlashGPT-6 Sol
Correct fields99.81%99.71%
Error-free documents98.67%98.00%
Per 1,000 documents$2.13$2.23
To fix / 1,0001320
Speed (median)2.3 sec2.9 sec
Valid answers100.00%100.00%

Free with login

Accuracy field by field and the documents where the models differ.

Just your email, no password.

Contract clauses · English (US) (150 documents)

GPT-6 Sol is more accurate by 0.1 pp. Gemini 3.8 Flash is 1.4× cheaper.

MetricGemini 3.8 FlashGPT-6 Sol
Correct fields99.90%100.00%
Error-free documents99.33%100.00%
Per 1,000 documents$2.47$3.33
To fix / 1,00070
Speed (median)2.0 sec2.5 sec
Valid answers100.00%100.00%

Free with login

Accuracy field by field and the documents where the models differ.

Just your email, no password.

Contract clauses · Czech (150 documents)

Both are equally accurate. Gemini 3.8 Flash is 1.5× cheaper.

MetricGemini 3.8 FlashGPT-6 Sol
Correct fields100.00%100.00%
Error-free documents100.00%100.00%
Per 1,000 documents$3.09$4.62
To fix / 1,00000
Speed (median)2.2 sec2.0 sec
Valid answers100.00%100.00%

Free with login

Accuracy field by field and the documents where the models differ.

Just your email, no password.

Contract clauses · German (150 documents)

GPT-6 Sol is more accurate by 0.1 pp. Gemini 3.8 Flash is 1.3× cheaper.

MetricGemini 3.8 FlashGPT-6 Sol
Correct fields99.90%100.00%
Error-free documents99.33%100.00%
Per 1,000 documents$3.42$4.36
To fix / 1,00070
Speed (median)2.6 sec2.9 sec
Valid answers100.00%100.00%

Free with login

Accuracy field by field and the documents where the models differ.

Just your email, no password.

Contract clauses · Spanish (150 documents)

Both are equally accurate. Gemini 3.8 Flash is 1.3× cheaper.

MetricGemini 3.8 FlashGPT-6 Sol
Correct fields100.00%100.00%
Error-free documents100.00%100.00%
Per 1,000 documents$3.33$4.48
To fix / 1,00000
Speed (median)3.0 sec2.8 sec
Valid answers100.00%100.00%

Free with login

Accuracy field by field and the documents where the models differ.

Just your email, no password.

Contract clauses · French (150 documents)

Gemini 3.8 Flash is more accurate by 0.4 pp. Gemini 3.8 Flash is 1.3× cheaper.

MetricGemini 3.8 FlashGPT-6 Sol
Correct fields100.00%99.59%
Error-free documents100.00%94.67%
Per 1,000 documents$3.50$4.49
To fix / 1,000053
Speed (median)3.0 sec2.6 sec
Valid answers100.00%100.00%

Free with login

Accuracy field by field and the documents where the models differ.

Just your email, no password.

Document summaries · English (US) (150 documents)

Gemini 3.8 Flash is more accurate by 0.4 pp. GPT-6 Sol is 1.4× cheaper.

MetricGemini 3.8 FlashGPT-6 Sol
Correct fields100.00%99.56%
Error-free documents100.00%97.33%
Per 1,000 documents$5.46$3.85
To fix / 1,000027
Speed (median)5.4 sec8.1 sec
Valid answers100.00%100.00%

Free with login

Accuracy field by field and the documents where the models differ.

Just your email, no password.

Document summaries · Czech (150 documents)

Both are equally accurate. GPT-6 Sol is 1.4× cheaper.

MetricGemini 3.8 FlashGPT-6 Sol
Correct fields100.00%100.00%
Error-free documents100.00%100.00%
Per 1,000 documents$6.69$4.72
To fix / 1,00000
Speed (median)6.2 sec7.6 sec
Valid answers100.00%100.00%

Free with login

Accuracy field by field and the documents where the models differ.

Just your email, no password.

Document summaries · German (150 documents)

Both are equally accurate. GPT-6 Sol is 1.5× cheaper.

MetricGemini 3.8 FlashGPT-6 Sol
Correct fields100.00%100.00%
Error-free documents100.00%100.00%
Per 1,000 documents$6.11$4.11
To fix / 1,00000
Speed (median)5.8 sec7.1 sec
Valid answers100.00%100.00%

Free with login

Accuracy field by field and the documents where the models differ.

Just your email, no password.

Document summaries · Spanish (150 documents)

Gemini 3.8 Flash is more accurate by 2.1 pp. GPT-6 Sol is 1.3× cheaper.

MetricGemini 3.8 FlashGPT-6 Sol
Correct fields100.00%97.89%
Error-free documents100.00%87.33%
Per 1,000 documents$6.78$5.02
To fix / 1,0000127
Speed (median)5.9 sec7.0 sec
Valid answers100.00%100.00%

Free with login

Accuracy field by field and the documents where the models differ.

Just your email, no password.

Document summaries · French (150 documents)

Gemini 3.8 Flash is more accurate by 2.8 pp. GPT-6 Sol is 1.5× cheaper.

MetricGemini 3.8 FlashGPT-6 Sol
Correct fields100.00%97.22%
Error-free documents100.00%83.33%
Per 1,000 documents$7.06$4.58
To fix / 1,0000167
Speed (median)6.7 sec6.1 sec
Valid answers100.00%100.00%

Free with login

Accuracy field by field and the documents where the models differ.

Just your email, no password.

Questions about a document · English (US) (150 documents)

GPT-6 Sol is more accurate by 0.1 pp. GPT-6 Sol is 1.3× cheaper.

MetricGemini 3.8 FlashGPT-6 Sol
Correct fields99.89%100.00%
Error-free documents99.33%100.00%
Per 1,000 documents$3.00$2.36
To fix / 1,00070
Speed (median)3.0 sec2.3 sec
Valid answers100.00%100.00%

Free with login

Accuracy field by field and the documents where the models differ.

Just your email, no password.

Questions about a document · Czech (150 documents)

Both are equally accurate. GPT-6 Sol is 1.1× cheaper.

MetricGemini 3.8 FlashGPT-6 Sol
Correct fields100.00%100.00%
Error-free documents100.00%100.00%
Per 1,000 documents$4.36$3.95
To fix / 1,00000
Speed (median)3.6 sec2.3 sec
Valid answers100.00%100.00%

Free with login

Accuracy field by field and the documents where the models differ.

Just your email, no password.

Questions about a document · German (150 documents)

Both are equally accurate. GPT-6 Sol is 1.2× cheaper.

MetricGemini 3.8 FlashGPT-6 Sol
Correct fields100.00%100.00%
Error-free documents100.00%100.00%
Per 1,000 documents$3.76$3.23
To fix / 1,00000
Speed (median)3.2 sec2.2 sec
Valid answers100.00%100.00%

Free with login

Accuracy field by field and the documents where the models differ.

Just your email, no password.

Questions about a document · Spanish (150 documents)

Both are equally accurate. GPT-6 Sol is 1.2× cheaper.

MetricGemini 3.8 FlashGPT-6 Sol
Correct fields100.00%100.00%
Error-free documents100.00%100.00%
Per 1,000 documents$4.05$3.36
To fix / 1,00000
Speed (median)3.6 sec2.5 sec
Valid answers100.00%100.00%

Free with login

Accuracy field by field and the documents where the models differ.

Just your email, no password.

Questions about a document · French (150 documents)

Both are equally accurate. GPT-6 Sol is 1.2× cheaper.

MetricGemini 3.8 FlashGPT-6 Sol
Correct fields100.00%100.00%
Error-free documents100.00%100.00%
Per 1,000 documents$4.06$3.43
To fix / 1,00000
Speed (median)3.7 sec2.5 sec
Valid answers100.00%100.00%

Free with login

Accuracy field by field and the documents where the models differ.

Just your email, no password.

Log in with email

We use your email only to log you in (and for the digest if you tick it). No spam.