Model comparison
DeepSeek-V3.1-Terminus vs Gemma 3 4B
DeepSeek-V3.1-Terminus is the stronger model overall, scoring 42.6 to 27.9 on the Noometry Index. Gemma 3 4B costs 9.5× less per token, which makes it the better buy when DeepSeek-V3.1-Terminus's lead doesn't matter for your workload.
Last verified . 13 shared benchmarks.
DeepSeek
GoogleSummary
- They share 13 benchmarks with published results for both. DeepSeek-V3.1-Terminus scores higher in 7 categories and Gemma 3 4B in 0 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where DeepSeek-V3.1-Terminus leads 35.7 to 15.8.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 57.4% for DeepSeek-V3.1-Terminus and 25.2% for Gemma 3 4B.
- Gemma 3 4B is cheaper at $0.04 / $0.08 per million input/output tokens, against $0.30 / $1 for DeepSeek-V3.1-Terminus.
- DeepSeek-V3.1-Terminus accepts more context: 164K tokens versus 131K.
Side by side
| DeepSeek-V3.1-Terminus | Gemma 3 4B | |
|---|---|---|
| Provider | DeepSeek | |
| Noometry Index | 42.6 | 27.9 |
| Released | 2025-09-22 | 2025-03-12 |
| Weights | Open | Open |
| Context window | 164K | 131K |
| Max output | 66K | 4K |
| Input $ / M tokens | $0.30 | $0.04 |
| Output $ / M tokens | $1 | $0.08 |
| Results tracked | 16 | 22 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 41.5 (#114), Gemma 3 4B: 35.3 (#214)
| Benchmark | DeepSeek-V3.1-Terminus | Gemma 3 4B |
|---|---|---|
| LMArena Coding | 1426 | 1230 |
| SciCode | 40.6% | — |
| ALE-Bench | 745.17 | — |
Agentic & Tool Use Not comparable
DeepSeek-V3.1-Terminus: —, Gemma 3 4B: 20.7 (#134)
| Benchmark | DeepSeek-V3.1-Terminus | Gemma 3 4B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 19.6% |
Reasoning DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 25.7 (#126), Gemma 3 4B: 12.7 (#330)
| Benchmark | DeepSeek-V3.1-Terminus | Gemma 3 4B |
|---|---|---|
| Kagi LLM Benchmark | 57.4% | 25.2% |
| LMArena Hard Prompts | 1427 | 1253 |
| DTBench | 81.3% | 50.9% |
| LMCA | 28.6% | 2.8% |
| CritPt | 1.7% | — |
| Chess Puzzles | — | 0% |
| Epoch Capabilities Index | — | 116.02 |
Math DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 35.7 (#138), Gemma 3 4B: 15.8 (#286)
| Benchmark | DeepSeek-V3.1-Terminus | Gemma 3 4B |
|---|---|---|
| LMArena Math | 1402 | 1239 |
| OTIS Mock AIME 2024-2025 | — | 7.5% |
Knowledge Not comparable
DeepSeek-V3.1-Terminus: —, Gemma 3 4B: 11.8 (#294)
| Benchmark | DeepSeek-V3.1-Terminus | Gemma 3 4B |
|---|---|---|
| GPQA Diamond | — | 23.2% |
| Vectara Hallucination Rate | — | 6.4% |
| LMArena Expert | — | 1223 |
Multilingual DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 52.3 (#87), Gemma 3 4B: 42.7 (#189)
| Benchmark | DeepSeek-V3.1-Terminus | Gemma 3 4B |
|---|---|---|
| LMArena Non-English | 1407 | 1273 |
| LMArena Russian | 1434 | 1294 |
| LMArena German | — | 1281 |
Instruction Following DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 74.4 (#96), Gemma 3 4B: 65.4 (#219)
| Benchmark | DeepSeek-V3.1-Terminus | Gemma 3 4B |
|---|---|---|
| LMArena Instruction Following | 1406 | 1239 |
Long Context DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 43.5 (#88), Gemma 3 4B: 38.6 (#189)
| Benchmark | DeepSeek-V3.1-Terminus | Gemma 3 4B |
|---|---|---|
| LMArena Longer Query | 1426 | 1274 |
Writing & Preference DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 61.3 (#87), Gemma 3 4B: 42.2 (#234)
| Benchmark | DeepSeek-V3.1-Terminus | Gemma 3 4B |
|---|---|---|
| LMArena Text | 1420 | 1291 |
| LMArena Creative Writing | 1404 | 1270 |
| LMArena Multi-Turn | 1411 | 1255 |
| EQ-Bench Creative Writing | — | 1068 |
Frequently asked questions
Is DeepSeek-V3.1-Terminus better than Gemma 3 4B?
DeepSeek-V3.1-Terminus is the stronger model overall, scoring 42.6 to 27.9 on the Noometry Index. Gemma 3 4B costs 9.5× less per token, which makes it the better buy when DeepSeek-V3.1-Terminus's lead doesn't matter for your workload.
Which is cheaper, DeepSeek-V3.1-Terminus or Gemma 3 4B?
Gemma 3 4B is cheaper. It lists at $0.04 per million input tokens and $0.08 per million output tokens; DeepSeek-V3.1-Terminus lists at $0.30 and $1.
Is DeepSeek-V3.1-Terminus or Gemma 3 4B better for coding?
DeepSeek-V3.1-Terminus scores higher on coding benchmarks: 41.5 versus 35.3 in the Noometry coding category.
Which has the bigger context window?
DeepSeek-V3.1-Terminus does, with 164K tokens against 131K.
How many benchmarks do DeepSeek-V3.1-Terminus and Gemma 3 4B share?
13 benchmarks have published results for both models. DeepSeek-V3.1-Terminus has 16 scored results on Noometry and Gemma 3 4B has 22.