Model comparison
DeepSeek-V2.5 (Sep 2024) vs Gemma 4 26B A4B IT
Gemma 4 26B A4B IT is the stronger model overall, scoring 43.5 to 37.6 on the Noometry Index.
Last verified . 14 shared benchmarks.
DeepSeek
GoogleSummary
- They share 14 benchmarks with published results for both. DeepSeek-V2.5 (Sep 2024) scores higher in 1 category and Gemma 4 26B A4B IT in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Gemma 4 26B A4B IT leads 47.6 to 35.9.
Side by side
| DeepSeek-V2.5 (Sep 2024) | Gemma 4 26B A4B IT | |
|---|---|---|
| Provider | DeepSeek | |
| Noometry Index | 37.6 | 43.5 |
| Released | 2024-09-06 | 2026-04-02 |
| Weights | Open | Open |
| Context window | — | 262K |
| Max output | — | 33K |
| Input $ / M tokens | — | $0.0675 |
| Output $ / M tokens | — | $0.23 |
| Results tracked | 22 | 28 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemma 4 26B A4B IT leads
DeepSeek-V2.5 (Sep 2024): 31.7 (#281), Gemma 4 26B A4B IT: 39.0 (#164)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemma 4 26B A4B IT |
|---|---|---|
| LMArena Coding | 1309 | 1447 |
| Aider Polyglot | 17.8% | — |
| LMArena WebDev | — | 1359 |
| SciCode | — | 40% |
| WeirdML | — | 35.2% |
| BigCodeBench Instruct | 48.6% | — |
| BigCodeBench Complete | 53.2% | — |
| ALE-Bench | — | 927.17 |
| HumanEval+ | 83.5% | — |
| MBPP+ | 74.1% | — |
Reasoning DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 25.6 (#145), Gemma 4 26B A4B IT: 21.8 (#213)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemma 4 26B A4B IT |
|---|---|---|
| LMArena Hard Prompts | 1289 | 1439 |
| CritPt | — | 0% |
| Chess Puzzles | — | 6% |
| DTBench | — | 74.9% |
| LMCA | — | 29.7% |
| Epoch Capabilities Index | — | 141.85 |
Math Gemma 4 26B A4B IT leads
DeepSeek-V2.5 (Sep 2024): 35.9 (#177), Gemma 4 26B A4B IT: 47.6 (#67)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemma 4 26B A4B IT |
|---|---|---|
| LMArena Math | 1288 | 1470 |
| OTIS Mock AIME 2024-2025 | — | 82.2% |
Knowledge Gemma 4 26B A4B IT leads
DeepSeek-V2.5 (Sep 2024): 34.8 (#193), Gemma 4 26B A4B IT: 45.7 (#85)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemma 4 26B A4B IT |
|---|---|---|
| LMArena Expert | 1266 | 1447 |
| GPQA Diamond | — | 73.2% |
| Vectara Hallucination Rate | — | 5.2% |
Multimodal Not comparable
DeepSeek-V2.5 (Sep 2024): —, Gemma 4 26B A4B IT: 40.6 (#46)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemma 4 26B A4B IT |
|---|---|---|
| LMArena Vision | — | 1260 |
Multilingual Gemma 4 26B A4B IT leads
DeepSeek-V2.5 (Sep 2024): 42.5 (#193), Gemma 4 26B A4B IT: 53.1 (#71)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemma 4 26B A4B IT |
|---|---|---|
| LMArena Non-English | 1273 | 1421 |
| LMArena Chinese | 1318 | 1495 |
| LMArena French | 1289 | 1460 |
| LMArena Russian | 1289 | 1434 |
| LMArena Spanish | 1248 | 1417 |
| LMArena German | 1258 | — |
| LMArena Japanese | 1228 | — |
| LMArena Korean | 1209 | — |
Instruction Following Gemma 4 26B A4B IT leads
DeepSeek-V2.5 (Sep 2024): 67.5 (#194), Gemma 4 26B A4B IT: 74.8 (#82)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemma 4 26B A4B IT |
|---|---|---|
| LMArena Instruction Following | 1280 | 1420 |
Long Context Gemma 4 26B A4B IT leads
DeepSeek-V2.5 (Sep 2024): 39.5 (#174), Gemma 4 26B A4B IT: 43.6 (#91)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemma 4 26B A4B IT |
|---|---|---|
| LMArena Longer Query | 1301 | 1428 |
Writing & Preference Gemma 4 26B A4B IT leads
DeepSeek-V2.5 (Sep 2024): 49.8 (#187), Gemma 4 26B A4B IT: 58.6 (#115)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemma 4 26B A4B IT |
|---|---|---|
| LMArena Text | 1294 | 1434 |
| LMArena Creative Writing | 1285 | 1402 |
| LMArena Multi-Turn | 1297 | 1441 |
| EQ-Bench Creative Writing | — | 1305 |
Frequently asked questions
Is DeepSeek-V2.5 (Sep 2024) better than Gemma 4 26B A4B IT?
Gemma 4 26B A4B IT is the stronger model overall, scoring 43.5 to 37.6 on the Noometry Index.
Is DeepSeek-V2.5 (Sep 2024) or Gemma 4 26B A4B IT better for coding?
Gemma 4 26B A4B IT scores higher on coding benchmarks: 39.0 versus 31.7 in the Noometry coding category.
How many benchmarks do DeepSeek-V2.5 (Sep 2024) and Gemma 4 26B A4B IT share?
14 benchmarks have published results for both models. DeepSeek-V2.5 (Sep 2024) has 22 scored results on Noometry and Gemma 4 26B A4B IT has 28.