Gemini 3.7 Flash vs Gemini 3.8 Flash
Arena Text · creative writing Arena Text · creative writing: Human pairwise preferences in the stated text vote pool
Gemini 3.7 Flash · high1,491
Gemini 3.8 Flash · high1,487
EQ-Bench Creative Writing v3 Creative Writing v3 · Elo: Judge-rated creative writing under the v3 prompt set and pairwise ranking
Gemini 3.7 Flash1,723.2
Gemini 3.8 Flash1,747.9
Lech Mazur Short-Story Lech Mazur Short-Story · comparison score: Judges compare stories written to ten required elements; the comparison score can be negative
Gemini 3.7 Flash · high-0.7
Gemini 3.8 Flash · high0.1
EQ-Bench Longform Creative Writing 1.11 Longform Creative Writing 1.11: Models plan, revise and write a story across eight turns, graded with a writing rubric
Gemini 3.7 Flash—
Gemini 3.8 Flash76.8
Harvey LAB-AA 1.1 Harvey LAB-AA 1.1: Legal agents work from case documents; every rubric criterion must pass and material hallucinations receive no credit
Gemini 3.7 Flash—
Gemini 3.8 Flash · high0
| Benchmark | Gemini 3.7 Flash | Gemini 3.8 Flash |
|---|---|---|
| Arena Text · creative writing | 1,491 · high | 1,487 · high |
| EQ-Bench Creative Writing v3 | 1,723.2 | 1,747.9 |
| Lech Mazur Short-Story | -0.7 · high | 0.1 · high |
| EQ-Bench Longform Creative Writing 1.11 | — | 76.8 |
| Harvey LAB-AA 1.1 | — | 0 · high |
Specs
| Spec | Gemini 3.7 Flash | Gemini 3.8 Flash |
|---|---|---|
| Input, per 1M | $0.75 | !$0.75 |
| Output, per 1M | $3.75 | !$3.75 |
| Cost per task? | — | $0.93 |
| Context | 1M | 1M |
| Weights | Closed | Closed |
| Released | Aug 2026 | Sep 2026 |
| See images | ✓ | ✓ |
| Read PDFs | Partial | ✓ |
| Hear audio | ✓ | ✓ |
| Watch video | ✓ | ✓ |
| Call tools | ✓ | ✓ |
| Force a tool call | Partial | Partial |
| Return strict JSON | ✓ | ✓ |
| Search the web | ? | ✓ |
| Run code | ? | ✓ |
| Use a computer | ? | ? |
What breaks when you switch
Gemini 3.7 Flash → Gemini 3.8 Flash
Sampling parameters are ignored›
temperature, top_p and top_k are ignored now and will cause HTTP 400 in future generations; remove them from requests.
No minimal thinking level›
The minimal thinking level returns an error; avoid minimal.
Gemini 3.8 Flash → Gemini 3.7 Flash
Nothing known.