ModelVerdict

Comparison

Gemini 3.8 Flash vs Gemma 4 31B

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

At a glance

Task · documentsEnglish (US)CzechGermanSpanishFrench
Invoice extraction99.8% / 97.6%100.0% / 96.1%97.2% / 94.9%95.9% / 95.4%99.9% / 93.6%
Email triage98.7% / 96.5%99.2% / 97.1%98.4% / 95.1%——
Personal data detection99.7% / 99.9%99.5% / 99.8%99.7% / 99.9%——
Contract clauses99.9% / 99.6%100.0% / 100.0%99.9% / 94.4%——

Correct fields: Gemini 3.8 Flash / Gemma 4 31B; the better one in bold. Click a cell to see the details.

Invoice extraction · English (US) (150 documents)

Gemini 3.8 Flash is more accurate by 2.2 pp. Gemma 4 31B is 22.9× cheaper.

MetricGemini 3.8 FlashGemma 4 31B
Correct fields99.8%97.6%
Error-free documents98%81%
Per 1,000 documents$3.84$0.17
To fix / 1,00020187
Speed (median)3.2 sec8.9 sec
Valid answers100.0%100.0%

Accuracy per field

FieldGemini 3.8 FlashGemma 4 31B
Supplier100.0%99.3%
Supplier ID (EIN, IČO…)100.0%98.0%
Invoice number100.0%100.0%
Payment reference100.0%99.3%
Invoice date100.0%98.7%
Due date100.0%98.7%
Net by tax rate100.0%95.3%
Tax by rate100.0%98.7%
Total100.0%100.0%
Currency100.0%100.0%
Bank account98.0%86.0%

Documents where they differ

Invoice extraction · Czech (150 documents)

Gemini 3.8 Flash is more accurate by 3.9 pp. Gemma 4 31B is 20.1× cheaper.

MetricGemini 3.8 FlashGemma 4 31B
Correct fields100.0%96.1%
Error-free documents99%67%
Per 1,000 documents$4.26$0.21
To fix / 1,0007333
Speed (median)3.5 sec10.4 sec
Valid answers100.0%100.0%

Accuracy per field

FieldGemini 3.8 FlashGemma 4 31B
Supplier99.3%96.7%
Supplier ID (EIN, IČO…)100.0%93.3%
VAT ID100.0%96.7%
Invoice number100.0%100.0%
Payment reference100.0%96.0%
Invoice date100.0%96.0%
Tax point100.0%95.3%
Due date100.0%94.0%
Net by tax rate100.0%98.0%
Tax by rate100.0%97.3%
Total100.0%99.3%
Currency100.0%100.0%
Bank account100.0%86.7%

Documents where they differ

Invoice extraction · German (150 documents)

Gemini 3.8 Flash is more accurate by 2.4 pp. Gemma 4 31B is 35.9× cheaper.

MetricGemini 3.8 FlashGemma 4 31B
Correct fields97.2%94.9%
Error-free documents96%64%
Per 1,000 documents$7.15$0.20
To fix / 1,00040360
Speed (median)4.7 sec7.7 sec
Valid answers97.3%100.0%

Accuracy per field

FieldGemini 3.8 FlashGemma 4 31B
Supplier97.3%98.7%
Supplier ID (EIN, IČO…)97.3%99.3%
VAT ID96.7%94.0%
Invoice number97.3%99.3%
Payment reference97.3%98.7%
Invoice date97.3%87.3%
Tax point97.3%92.0%
Due date97.3%90.7%
Net by tax rate96.7%94.7%
Tax by rate97.3%97.3%
Total97.3%99.3%
Currency97.3%99.3%
Bank account97.3%82.7%

Documents where they differ

Invoice extraction · Spanish (100 documents)

Gemini 3.8 Flash is more accurate by 0.5 pp. Gemma 4 31B is 38.6× cheaper.

MetricGemini 3.8 FlashGemma 4 31B
Correct fields95.9%95.4%
Error-free documents95%54%
Per 1,000 documents$7.86$0.20
To fix / 1,00050460
Speed (median)5.3 sec8.7 sec
Valid answers96.0%100.0%

Accuracy per field

FieldGemini 3.8 FlashGemma 4 31B
Supplier95.0%95.4%
Supplier ID (EIN, IČO…)96.0%93.1%
VAT ID96.0%98.9%
Invoice number96.0%100.0%
Payment reference96.0%95.4%
Invoice date96.0%98.9%
Tax point96.0%100.0%
Due date96.0%97.7%
Net by tax rate96.0%81.6%
Tax by rate96.0%100.0%
Total96.0%100.0%
Currency96.0%100.0%
Bank account96.0%79.3%

Documents where they differ

Invoice extraction · French (100 documents)

Gemini 3.8 Flash is more accurate by 6.4 pp. Gemma 4 31B is 27.8× cheaper.

MetricGemini 3.8 FlashGemma 4 31B
Correct fields99.9%93.6%
Error-free documents99%51%
Per 1,000 documents$6.18$0.22
To fix / 1,00010494
Speed (median)5.2 sec7.2 sec
Valid answers100.0%100.0%

Accuracy per field

FieldGemini 3.8 FlashGemma 4 31B
Supplier100.0%96.6%
Supplier ID (EIN, IČO…)100.0%77.0%
VAT ID100.0%95.4%
Invoice number100.0%97.7%
Payment reference99.0%96.6%
Invoice date100.0%96.6%
Tax point100.0%94.3%
Due date100.0%96.6%
Net by tax rate100.0%93.1%
Tax by rate100.0%100.0%
Total100.0%100.0%
Currency100.0%98.9%
Bank account100.0%73.6%

Documents where they differ

Email triage · English (US) (150 documents)

Gemini 3.8 Flash is more accurate by 2.2 pp. Gemma 4 31B is 19.8× cheaper.

MetricGemini 3.8 FlashGemma 4 31B
Correct fields98.7%96.5%
Error-free documents89%72%
Per 1,000 documents$2.39$0.12
To fix / 1,000113280
Speed (median)2.2 sec2.2 sec
Valid answers100.0%100.0%

Accuracy per field

FieldGemini 3.8 FlashGemma 4 31B
Category100.0%100.0%
Priority100.0%85.3%
Sentiment88.7%83.3%
Customer name100.0%100.0%
Order number100.0%100.0%
Invoice number100.0%100.0%
Amount100.0%100.0%
Currency100.0%100.0%
Deadline100.0%100.0%

Documents where they differ

Email triage · Czech (150 documents)

Gemini 3.8 Flash is more accurate by 2.1 pp. Gemma 4 31B is 19.3× cheaper.

MetricGemini 3.8 FlashGemma 4 31B
Correct fields99.2%97.1%
Error-free documents93%75%
Per 1,000 documents$3.06$0.16
To fix / 1,00073253
Speed (median)2.6 sec2.5 sec
Valid answers100.0%100.0%

Accuracy per field

FieldGemini 3.8 FlashGemma 4 31B
Category100.0%100.0%
Priority99.3%85.3%
Sentiment93.3%88.7%
Customer name100.0%100.0%
Order number100.0%100.0%
Invoice number100.0%100.0%
Amount100.0%100.0%
Currency100.0%100.0%
Deadline100.0%100.0%

Documents where they differ

Email triage · German (150 documents)

Gemini 3.8 Flash is more accurate by 3.3 pp. Gemma 4 31B is 18.5× cheaper.

MetricGemini 3.8 FlashGemma 4 31B
Correct fields98.4%95.1%
Error-free documents85%62%
Per 1,000 documents$2.74$0.15
To fix / 1,000147380
Speed (median)2.9 sec3.6 sec
Valid answers100.0%100.0%

Accuracy per field

FieldGemini 3.8 FlashGemma 4 31B
Category100.0%100.0%
Priority100.0%78.0%
Sentiment85.3%78.0%
Customer name100.0%100.0%
Order number100.0%100.0%
Invoice number100.0%100.0%
Amount100.0%100.0%
Currency100.0%100.0%
Deadline100.0%100.0%

Documents where they differ

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

Gemma 4 31B is more accurate by 0.2 pp. Gemma 4 31B is 22.1× cheaper.

MetricGemini 3.8 FlashGemma 4 31B
Correct fields99.7%99.9%
Error-free documents98%99%
Per 1,000 documents$1.86$0.08
To fix / 1,000207
Speed (median)1.9 sec2.4 sec
Valid answers100.0%100.0%

Accuracy per field

FieldGemini 3.8 FlashGemma 4 31B
Names100.0%100.0%
Email addresses100.0%100.0%
Phone numbers100.0%100.0%
Addresses98.0%99.3%
Dates of birth100.0%100.0%
National IDs100.0%100.0%
Bank accounts100.0%100.0%

Documents where they differ

Both models got the same number of fields right on every example document.

Personal data detection · Czech (150 documents)

Gemma 4 31B is more accurate by 0.3 pp. Gemma 4 31B is 18.4× cheaper.

MetricGemini 3.8 FlashGemma 4 31B
Correct fields99.5%99.8%
Error-free documents97%99%
Per 1,000 documents$2.19$0.12
To fix / 1,0003313
Speed (median)2.0 sec2.9 sec
Valid answers100.0%100.0%

Accuracy per field

FieldGemini 3.8 FlashGemma 4 31B
Names98.7%100.0%
Email addresses100.0%100.0%
Phone numbers100.0%100.0%
Addresses98.0%98.7%
Dates of birth100.0%100.0%
National IDs100.0%100.0%
Bank accounts100.0%100.0%

Documents where they differ

Both models got the same number of fields right on every example document.

Personal data detection · German (150 documents)

Gemma 4 31B is more accurate by 0.2 pp. Gemma 4 31B is 20.8× cheaper.

MetricGemini 3.8 FlashGemma 4 31B
Correct fields99.7%99.9%
Error-free documents98%99%
Per 1,000 documents$2.12$0.10
To fix / 1,000207
Speed (median)2.3 sec1.9 sec
Valid answers100.0%100.0%

Accuracy per field

FieldGemini 3.8 FlashGemma 4 31B
Names100.0%100.0%
Email addresses100.0%100.0%
Phone numbers100.0%100.0%
Addresses98.0%99.3%
Dates of birth100.0%100.0%
National IDs100.0%100.0%
Bank accounts100.0%100.0%

Documents where they differ

Contract clauses · English (US) (150 documents)

Gemini 3.8 Flash is more accurate by 0.3 pp. Gemma 4 31B is 13.7× cheaper.

MetricGemini 3.8 FlashGemma 4 31B
Correct fields99.9%99.6%
Error-free documents99%97%
Per 1,000 documents$2.47$0.18
To fix / 1,000727
Speed (median)2.0 sec2.5 sec
Valid answers100.0%100.0%

Accuracy per field

FieldGemini 3.8 FlashGemma 4 31B
Provider99.3%98.0%
Provider ID100.0%100.0%
Client99.3%98.0%
Client ID100.0%100.0%
Contract type100.0%100.0%
Price (net)100.0%100.0%
Currency100.0%100.0%
Billing period100.0%100.0%
Effective date100.0%99.3%
End date100.0%100.0%
Notice period (months)100.0%100.0%
Late-payment penalty (% per day)100.0%100.0%
Governing law100.0%100.0%

Documents where they differ

Contract clauses · Czech (150 documents)

Both are equally accurate. Gemma 4 31B is 12.2× cheaper.

MetricGemini 3.8 FlashGemma 4 31B
Correct fields100.0%100.0%
Error-free documents100%99%
Per 1,000 documents$3.09$0.25
To fix / 1,00007
Speed (median)2.2 sec6.0 sec
Valid answers100.0%100.0%

Accuracy per field

FieldGemini 3.8 FlashGemma 4 31B
Provider100.0%100.0%
Provider ID100.0%100.0%
Client100.0%100.0%
Client ID100.0%100.0%
Contract type100.0%100.0%
Price (net)100.0%100.0%
Currency100.0%100.0%
Billing period100.0%100.0%
Effective date100.0%99.3%
End date100.0%100.0%
Notice period (months)100.0%100.0%
Late-payment penalty (% per day)100.0%100.0%
Governing law100.0%100.0%

Documents where they differ

Contract clauses · German (150 documents)

Gemini 3.8 Flash is more accurate by 5.5 pp. Gemma 4 31B is 15.5× cheaper.

MetricGemini 3.8 FlashGemma 4 31B
Correct fields99.9%94.4%
Error-free documents99%27%
Per 1,000 documents$3.42$0.22
To fix / 1,0007727
Speed (median)2.6 sec2.6 sec
Valid answers100.0%100.0%

Accuracy per field

FieldGemini 3.8 FlashGemma 4 31B
Provider100.0%100.0%
Provider ID99.3%100.0%
Client100.0%100.0%
Client ID99.3%100.0%
Contract type100.0%100.0%
Price (net)100.0%100.0%
Currency100.0%100.0%
Billing period100.0%100.0%
Effective date100.0%100.0%
End date100.0%100.0%
Notice period (months)100.0%100.0%
Late-payment penalty (% per day)100.0%27.3%
Governing law100.0%100.0%

Documents where they differ