ModelVerdict

Comparison

GPT-6 Luna vs Mistral Medium 3.5

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

Verdict

GPT-6 Luna is more accurate and cheaper in 30 of 30 combinations.

More accurate

GPT-6 Luna

wins 13 of 30

Cheaper

GPT-6 Luna

cheaper in 30 of 30

Faster

Mistral Medium 3.5

1.0 sec vs 3.0 sec

Fewer fixes

Mistral Medium 3.5

20 vs 33 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 · documentsGPT-6 LunaMistral Medium 3.5DifferenceGPT-6 LunaMistral Medium 3.5Difference
Invoice extraction99.70%97.76%▼ 1.94$0.51$4.37+762 %
Email triage97.48%94.89%▼ 2.59$0.13$1.47+994 %
Personal data detection99.62%99.71%▲ 0.09$0.11$1.15+984 %
Contract clauses99.79%98.10%▼ 1.69$0.20$2.35+1069 %
Document summaries100.00%100.00%±0$0.16$1.57+882 %
Questions about a document99.11%100.00%▲ 0.89$0.15$1.81+1086 %
Czech
Accuracy (correct fields)Per 1,000 documents
Task · documentsGPT-6 LunaMistral Medium 3.5DifferenceGPT-6 LunaMistral Medium 3.5Difference
Invoice extraction99.28%93.49%▼ 5.79$0.51$4.69+828 %
Email triage97.93%94.52%▼ 3.41$0.18$1.95+993 %
Personal data detection99.62%99.81%▲ 0.19$0.12$1.40+1070 %
Contract clauses100.00%100.00%±0$0.27$3.06+1032 %
Document summaries100.00%100.00%±0$0.21$2.31+988 %
Questions about a document100.00%100.00%±0$0.23$2.51+1004 %
German
Accuracy (correct fields)Per 1,000 documents
Task · documentsGPT-6 LunaMistral Medium 3.5DifferenceGPT-6 LunaMistral Medium 3.5Difference
Invoice extraction99.23%97.38%▼ 1.85$0.54$4.72+767 %
Email triage96.96%92.74%▼ 4.22$0.15$1.77+1062 %
Personal data detection99.62%100.00%▲ 0.38$0.12$1.38+1080 %
Contract clauses100.00%97.13%▼ 2.87$0.27$2.76+937 %
Document summaries100.00%100.00%±0$0.19$1.98+919 %
Questions about a document98.00%100.00%▲ 2$0.20$2.13+961 %
Spanish
Accuracy (correct fields)Per 1,000 documents
Task · documentsGPT-6 LunaMistral Medium 3.5DifferenceGPT-6 LunaMistral Medium 3.5Difference
Invoice extraction99.13%96.87%▼ 2.26$0.55$4.75+760 %
Email triage96.81%94.15%▼ 2.66$0.16$1.79+1031 %
Personal data detection99.71%99.62%▼ 0.09$0.12$1.41+1088 %
Contract clauses100.00%100.00%±0$0.26$3.02+1042 %
Document summaries97.00%100.00%▲ 3$0.19$1.86+884 %
Questions about a document99.67%100.00%▲ 0.33$0.19$2.18+1020 %
French
Accuracy (correct fields)Per 1,000 documents
Task · documentsGPT-6 LunaMistral Medium 3.5DifferenceGPT-6 LunaMistral Medium 3.5Difference
Invoice extraction99.08%98.56%▼ 0.52$0.56$4.82+763 %
Email triage97.26%94.15%▼ 3.11$0.16$1.78+1011 %
Personal data detection99.62%99.62%±0$0.12$1.43+1054 %
Contract clauses99.33%100.00%▲ 0.67$0.27$3.01+1001 %
Document summaries98.78%100.00%▲ 1.22$0.20$1.87+851 %
Questions about a document99.56%100.00%▲ 0.44$0.20$2.19+973 %

Difference: Mistral Medium 3.5 vs. GPT-6 Luna; 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 Luna is more accurate by 1.9 pp. GPT-6 Luna is 8.6× cheaper.

MetricGPT-6 LunaMistral Medium 3.5
Correct fields99.70%97.76%
Error-free documents96.67%79.33%
Per 1,000 documents$0.51$4.37
To fix / 1,00033207
Speed (median)3.4 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.

Invoice extraction · Czech (150 documents)

GPT-6 Luna is more accurate by 5.8 pp. GPT-6 Luna is 9.3× cheaper.

MetricGPT-6 LunaMistral Medium 3.5
Correct fields99.28%93.49%
Error-free documents93.33%63.33%
Per 1,000 documents$0.51$4.69
To fix / 1,00067367
Speed (median)3.2 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.

Invoice extraction · German (150 documents)

GPT-6 Luna is more accurate by 1.8 pp. GPT-6 Luna is 8.7× cheaper.

MetricGPT-6 LunaMistral Medium 3.5
Correct fields99.23%97.38%
Error-free documents91.33%74.67%
Per 1,000 documents$0.54$4.72
To fix / 1,00087253
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.

Invoice extraction · Spanish (150 documents)

GPT-6 Luna is more accurate by 2.3 pp. GPT-6 Luna is 8.6× cheaper.

MetricGPT-6 LunaMistral Medium 3.5
Correct fields99.13%96.87%
Error-free documents88.67%66.67%
Per 1,000 documents$0.55$4.75
To fix / 1,000113333
Speed (median)3.9 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.

Invoice extraction · French (150 documents)

GPT-6 Luna is more accurate by 0.5 pp. GPT-6 Luna is 8.6× cheaper.

MetricGPT-6 LunaMistral Medium 3.5
Correct fields99.08%98.56%
Error-free documents88.00%84.67%
Per 1,000 documents$0.56$4.82
To fix / 1,000120153
Speed (median)3.9 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.

Email triage · English (US) (150 documents)

GPT-6 Luna is more accurate by 2.6 pp. GPT-6 Luna is 10.9× cheaper.

MetricGPT-6 LunaMistral Medium 3.5
Correct fields97.48%94.89%
Error-free documents77.33%57.33%
Per 1,000 documents$0.13$1.47
To fix / 1,000227427
Speed (median)2.5 sec0.8 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)

GPT-6 Luna is more accurate by 3.4 pp. GPT-6 Luna is 10.9× cheaper.

MetricGPT-6 LunaMistral Medium 3.5
Correct fields97.93%94.52%
Error-free documents81.33%55.33%
Per 1,000 documents$0.18$1.95
To fix / 1,000187447
Speed (median)2.6 sec0.8 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)

GPT-6 Luna is more accurate by 4.2 pp. GPT-6 Luna is 11.6× cheaper.

MetricGPT-6 LunaMistral Medium 3.5
Correct fields96.96%92.74%
Error-free documents72.67%48.00%
Per 1,000 documents$0.15$1.77
To fix / 1,000273520
Speed (median)2.8 sec0.8 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 · Spanish (150 documents)

GPT-6 Luna is more accurate by 2.7 pp. GPT-6 Luna is 11.3× cheaper.

MetricGPT-6 LunaMistral Medium 3.5
Correct fields96.81%94.15%
Error-free documents72.00%54.00%
Per 1,000 documents$0.16$1.79
To fix / 1,000280460
Speed (median)3.1 sec0.9 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 · French (150 documents)

GPT-6 Luna is more accurate by 3.1 pp. GPT-6 Luna is 11.1× cheaper.

MetricGPT-6 LunaMistral Medium 3.5
Correct fields97.26%94.15%
Error-free documents76.00%54.00%
Per 1,000 documents$0.16$1.78
To fix / 1,000240460
Speed (median)2.8 sec0.9 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 · English (US) (150 documents)

Mistral Medium 3.5 is more accurate by 0.1 pp. GPT-6 Luna is 10.8× cheaper.

MetricGPT-6 LunaMistral Medium 3.5
Correct fields99.62%99.71%
Error-free documents97.33%98.00%
Per 1,000 documents$0.11$1.15
To fix / 1,0002720
Speed (median)1.8 sec0.8 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.

Personal data detection · Czech (150 documents)

Mistral Medium 3.5 is more accurate by 0.2 pp. GPT-6 Luna is 11.7× cheaper.

MetricGPT-6 LunaMistral Medium 3.5
Correct fields99.62%99.81%
Error-free documents97.33%98.67%
Per 1,000 documents$0.12$1.40
To fix / 1,0002713
Speed (median)1.8 sec0.8 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.

Personal data detection · German (150 documents)

Mistral Medium 3.5 is more accurate by 0.4 pp. GPT-6 Luna is 11.8× cheaper.

MetricGPT-6 LunaMistral Medium 3.5
Correct fields99.62%100.00%
Error-free documents97.33%100.00%
Per 1,000 documents$0.12$1.38
To fix / 1,000270
Speed (median)2.2 sec0.9 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 · Spanish (150 documents)

GPT-6 Luna is more accurate by 0.1 pp. GPT-6 Luna is 11.9× cheaper.

MetricGPT-6 LunaMistral Medium 3.5
Correct fields99.71%99.62%
Error-free documents98.00%97.33%
Per 1,000 documents$0.12$1.41
To fix / 1,0002027
Speed (median)2.3 sec1.0 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.

Personal data detection · French (150 documents)

Both are equally accurate. GPT-6 Luna is 11.5× cheaper.

MetricGPT-6 LunaMistral Medium 3.5
Correct fields99.62%99.62%
Error-free documents97.33%97.33%
Per 1,000 documents$0.12$1.43
To fix / 1,0002727
Speed (median)2.4 sec0.9 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 Luna is more accurate by 1.7 pp. GPT-6 Luna is 11.7× cheaper.

MetricGPT-6 LunaMistral Medium 3.5
Correct fields99.79%98.10%
Error-free documents98.67%87.33%
Per 1,000 documents$0.20$2.35
To fix / 1,00013127
Speed (median)2.7 sec1.1 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.

Contract clauses · Czech (150 documents)

Both are equally accurate. GPT-6 Luna is 11.3× cheaper.

MetricGPT-6 LunaMistral Medium 3.5
Correct fields100.00%100.00%
Error-free documents100.00%100.00%
Per 1,000 documents$0.27$3.06
To fix / 1,00000
Speed (median)2.6 sec1.2 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.

Contract clauses · German (150 documents)

GPT-6 Luna is more accurate by 2.9 pp. GPT-6 Luna is 10.4× cheaper.

MetricGPT-6 LunaMistral Medium 3.5
Correct fields100.00%97.13%
Error-free documents100.00%75.33%
Per 1,000 documents$0.27$2.76
To fix / 1,0000247
Speed (median)3.0 sec1.2 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.

Contract clauses · Spanish (150 documents)

Both are equally accurate. GPT-6 Luna is 11.4× cheaper.

MetricGPT-6 LunaMistral Medium 3.5
Correct fields100.00%100.00%
Error-free documents100.00%100.00%
Per 1,000 documents$0.26$3.02
To fix / 1,00000
Speed (median)3.1 sec1.3 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.

Contract clauses · French (150 documents)

Mistral Medium 3.5 is more accurate by 0.7 pp. GPT-6 Luna is 11× cheaper.

MetricGPT-6 LunaMistral Medium 3.5
Correct fields99.33%100.00%
Error-free documents91.33%100.00%
Per 1,000 documents$0.27$3.01
To fix / 1,000870
Speed (median)3.2 sec1.3 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.

Document summaries · English (US) (150 documents)

Both are equally accurate. GPT-6 Luna is 9.8× cheaper.

MetricGPT-6 LunaMistral Medium 3.5
Correct fields100.00%100.00%
Error-free documents100.00%100.00%
Per 1,000 documents$0.16$1.57
To fix / 1,00000
Speed (median)2.9 sec0.9 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.

Document summaries · Czech (150 documents)

Both are equally accurate. GPT-6 Luna is 10.9× cheaper.

MetricGPT-6 LunaMistral Medium 3.5
Correct fields100.00%100.00%
Error-free documents100.00%100.00%
Per 1,000 documents$0.21$2.31
To fix / 1,00000
Speed (median)3.4 sec1.4 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 Luna is 10.2× cheaper.

MetricGPT-6 LunaMistral Medium 3.5
Correct fields100.00%100.00%
Error-free documents100.00%100.00%
Per 1,000 documents$0.19$1.98
To fix / 1,00000
Speed (median)3.3 sec1.1 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)

Mistral Medium 3.5 is more accurate by 3 pp. GPT-6 Luna is 9.8× cheaper.

MetricGPT-6 LunaMistral Medium 3.5
Correct fields97.00%100.00%
Error-free documents82.00%100.00%
Per 1,000 documents$0.19$1.86
To fix / 1,0001800
Speed (median)3.0 sec1.0 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)

Mistral Medium 3.5 is more accurate by 1.2 pp. GPT-6 Luna is 9.5× cheaper.

MetricGPT-6 LunaMistral Medium 3.5
Correct fields98.78%100.00%
Error-free documents92.67%100.00%
Per 1,000 documents$0.20$1.87
To fix / 1,000730
Speed (median)3.2 sec1.0 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.

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

Mistral Medium 3.5 is more accurate by 0.9 pp. GPT-6 Luna is 11.9× cheaper.

MetricGPT-6 LunaMistral Medium 3.5
Correct fields99.11%100.00%
Error-free documents96.00%100.00%
Per 1,000 documents$0.15$1.81
To fix / 1,000400
Speed (median)3.1 sec0.8 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.

Questions about a document · Czech (150 documents)

Both are equally accurate. GPT-6 Luna is 11× cheaper.

MetricGPT-6 LunaMistral Medium 3.5
Correct fields100.00%100.00%
Error-free documents100.00%100.00%
Per 1,000 documents$0.23$2.51
To fix / 1,00000
Speed (median)3.1 sec0.9 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.

Questions about a document · German (150 documents)

Mistral Medium 3.5 is more accurate by 2 pp. GPT-6 Luna is 10.6× cheaper.

MetricGPT-6 LunaMistral Medium 3.5
Correct fields98.00%100.00%
Error-free documents91.33%100.00%
Per 1,000 documents$0.20$2.13
To fix / 1,000870
Speed (median)3.2 sec0.8 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.

Questions about a document · Spanish (150 documents)

Mistral Medium 3.5 is more accurate by 0.3 pp. GPT-6 Luna is 11.2× cheaper.

MetricGPT-6 LunaMistral Medium 3.5
Correct fields99.67%100.00%
Error-free documents98.67%100.00%
Per 1,000 documents$0.19$2.18
To fix / 1,000130
Speed (median)2.9 sec0.8 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.

Questions about a document · French (150 documents)

Mistral Medium 3.5 is more accurate by 0.4 pp. GPT-6 Luna is 10.7× cheaper.

MetricGPT-6 LunaMistral Medium 3.5
Correct fields99.56%100.00%
Error-free documents98.00%100.00%
Per 1,000 documents$0.20$2.19
To fix / 1,000200
Speed (median)3.0 sec0.9 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.

Log in with email

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