Model comparison
Gemini 3 Deep Think vs Step 3.5 Flash
Step 3.5 Flash has enough public results to be ranked (#116); Gemini 3 Deep Think does not yet, so treat this comparison as directional.
Last verified . 0 shared benchmarks.
Google
StepFunSummary
- The widest gap is in reasoning, where Gemini 3 Deep Think leads 70.2 to 22.2.
- Step 3.5 Flash has downloadable open weights; the other is API-only.
Side by side
| Gemini 3 Deep Think | Step 3.5 Flash | |
|---|---|---|
| Provider | StepFun | |
| Noometry Index | 58.3 | 42.3 |
| Released | 2026-02-12 | 2026-01-29 |
| Weights | Proprietary | Open |
| Context window | — | 256K |
| Max output | — | 256K |
| Input $ / M tokens | — | $0.10 |
| Output $ / M tokens | — | $0.30 |
| Results tracked | 3 | 19 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Gemini 3 Deep Think: —, Step 3.5 Flash: 42.4 (#105)
| Benchmark | Gemini 3 Deep Think | Step 3.5 Flash |
|---|---|---|
| LMArena Coding | — | 1436 |
Reasoning Gemini 3 Deep Think leads
Gemini 3 Deep Think: 70.2 (#14), Step 3.5 Flash: 22.2 (#202)
| Benchmark | Gemini 3 Deep Think | Step 3.5 Flash |
|---|---|---|
| ARC-AGI-2 | 84.6% | — |
| NYT Connections (extended) | — | 28.4% |
| ARC-AGI-1 | 96% | — |
| CritPt | 25.7% | — |
| LMArena Hard Prompts | — | 1411 |
Math Not comparable
Gemini 3 Deep Think: —, Step 3.5 Flash: 42.6 (#84)
| Benchmark | Gemini 3 Deep Think | Step 3.5 Flash |
|---|---|---|
| MathArena Final-Answer Competitions | — | 66.8% |
| LMArena Math | — | 1408 |
Knowledge Not comparable
Gemini 3 Deep Think: —, Step 3.5 Flash: 39.6 (#132)
| Benchmark | Gemini 3 Deep Think | Step 3.5 Flash |
|---|---|---|
| LMArena Expert | — | 1421 |
Multilingual Not comparable
Gemini 3 Deep Think: —, Step 3.5 Flash: 50.5 (#119)
| Benchmark | Gemini 3 Deep Think | Step 3.5 Flash |
|---|---|---|
| LMArena Non-English | — | 1385 |
| LMArena Chinese | — | 1447 |
| LMArena French | — | 1421 |
| LMArena German | — | 1405 |
| LMArena Japanese | — | 1354 |
| LMArena Korean | — | 1352 |
| LMArena Russian | — | 1385 |
| LMArena Spanish | — | 1419 |
Instruction Following Not comparable
Gemini 3 Deep Think: —, Step 3.5 Flash: 73.1 (#124)
| Benchmark | Gemini 3 Deep Think | Step 3.5 Flash |
|---|---|---|
| LMArena Instruction Following | — | 1385 |
Long Context Not comparable
Gemini 3 Deep Think: —, Step 3.5 Flash: 42.8 (#117)
| Benchmark | Gemini 3 Deep Think | Step 3.5 Flash |
|---|---|---|
| LMArena Longer Query | — | 1402 |
Writing & Preference Not comparable
Gemini 3 Deep Think: —, Step 3.5 Flash: 58.8 (#113)
| Benchmark | Gemini 3 Deep Think | Step 3.5 Flash |
|---|---|---|
| LMArena Text | — | 1403 |
| LMArena Creative Writing | — | 1357 |
| LMArena Multi-Turn | — | 1405 |
Frequently asked questions
Is Gemini 3 Deep Think better than Step 3.5 Flash?
Step 3.5 Flash has enough public results to be ranked (#116); Gemini 3 Deep Think does not yet, so treat this comparison as directional.
How many benchmarks do Gemini 3 Deep Think and Step 3.5 Flash share?
0 benchmarks have published results for both models. Gemini 3 Deep Think has 3 scored results on Noometry and Step 3.5 Flash has 19.