Model comparison
DeepSeek-V3.2-Exp vs Gemma 4 26B A4B IT
DeepSeek-V3.2-Exp is the stronger model overall, scoring 44.9 to 43.1 on the Noometry Index. Gemma 4 26B A4B IT costs 2.4× less per token, which makes it the better buy when DeepSeek-V3.2-Exp's lead doesn't matter for your workload.
Last verified . 26 shared benchmarks.
DeepSeek
GoogleSummary
- They share 26 benchmarks with published results for both. DeepSeek-V3.2-Exp scores higher in 5 categories and Gemma 4 26B A4B IT in 3 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in coding, where DeepSeek-V3.2-Exp leads 46.4 to 38.8.
- The biggest single-benchmark swing is DTBench: 87.7% for DeepSeek-V3.2-Exp and 74.9% for Gemma 4 26B A4B IT.
- Gemma 4 26B A4B IT is cheaper at $0.0765 / $0.26 per million input/output tokens, against $0.26 / $0.38 for DeepSeek-V3.2-Exp.
- Gemma 4 26B A4B IT accepts more context: 262K tokens versus 164K.
Side by side
| DeepSeek-V3.2-Exp | Gemma 4 26B A4B IT | |
|---|---|---|
| Provider | DeepSeek | |
| Noometry Index | 44.9 | 43.1 |
| Released | 2025-09-29 | 2026-04-02 |
| Weights | Open | Open |
| Context window | 164K | 262K |
| Max output | 66K | 33K |
| Input $ / M tokens | $0.26 | $0.0765 |
| Output $ / M tokens | $0.38 | $0.26 |
| Results tracked | 48 | 28 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 46.4 (#61), Gemma 4 26B A4B IT: 38.8 (#156)
| Benchmark | DeepSeek-V3.2-Exp | Gemma 4 26B A4B IT |
|---|---|---|
| LMArena WebDev | 1362 | 1359 |
| SciCode | 38.9% | 40% |
| WeirdML | 39.5% | 35.2% |
| LMArena Coding | 1454 | 1446 |
| SWE-bench Verified (bash only) | 70% | — |
| Aider Polyglot | 74.2% | — |
| SWE-bench Multilingual | 59% | — |
| ALE-Bench | — | 927.17 |
Agentic & Tool Use Not comparable
DeepSeek-V3.2-Exp: 37.9 (#31), Gemma 4 26B A4B IT: —
| Benchmark | DeepSeek-V3.2-Exp | Gemma 4 26B A4B IT |
|---|---|---|
| Terminal-Bench | 39.6% | — |
| Berkeley Function Calling Leaderboard | 56.7% | — |
| TheAgentCompany | 42.9% | — |
| Vending-Bench 2 | 1,034 | — |
Reasoning Too close to call
DeepSeek-V3.2-Exp: 22.0 (#185), Gemma 4 26B A4B IT: 21.1 (#205)
| Benchmark | DeepSeek-V3.2-Exp | Gemma 4 26B A4B IT |
|---|---|---|
| CritPt | 2.9% | 0% |
| Chess Puzzles | 14% | 6% |
| LMArena Hard Prompts | 1433 | 1438 |
| DTBench | 87.7% | 74.9% |
| LMCA | 29.1% | 29.7% |
| Epoch Capabilities Index | 146.28 | 141.85 |
| ARC-AGI-2 | 4% | — |
| Kagi LLM Benchmark | 52.2% | — |
| NYT Connections (extended) | 36.7% | — |
| ARC-AGI-1 | 57% | — |
| Thematic Generalization | 65% | — |
Math Gemma 4 26B A4B IT leads
DeepSeek-V3.2-Exp: 40.8 (#75), Gemma 4 26B A4B IT: 45.7 (#61)
| Benchmark | DeepSeek-V3.2-Exp | Gemma 4 26B A4B IT |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 87.8% | 82.2% |
| LMArena Math | 1435 | 1469 |
| MathArena Final-Answer Competitions | 57.7% | — |
| ProofBench | 8% | — |
| FrontierMath (Feb 2025 set) | 22.1% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 51.6 (#61), Gemma 4 26B A4B IT: 45.6 (#79)
| Benchmark | DeepSeek-V3.2-Exp | Gemma 4 26B A4B IT |
|---|---|---|
| GPQA Diamond | 83.4% | 73.2% |
| Vectara Hallucination Rate | 5.3% | 5.2% |
| LMArena Expert | 1436 | 1447 |
Multimodal Not comparable
DeepSeek-V3.2-Exp: —, Gemma 4 26B A4B IT: 39.9 (#43)
| Benchmark | DeepSeek-V3.2-Exp | Gemma 4 26B A4B IT |
|---|---|---|
| LMArena Vision | — | 1260 |
Multilingual Too close to call
DeepSeek-V3.2-Exp: 52.4 (#85), Gemma 4 26B A4B IT: 53.3 (#68)
| Benchmark | DeepSeek-V3.2-Exp | Gemma 4 26B A4B IT |
|---|---|---|
| LMArena Non-English | 1409 | 1422 |
| LMArena Chinese | 1461 | 1495 |
| LMArena French | 1435 | 1459 |
| LMArena Russian | 1423 | 1434 |
| LMArena Spanish | 1441 | 1415 |
| LMArena German | 1440 | — |
| LMArena Japanese | 1377 | — |
| LMArena Korean | 1372 | — |
Instruction Following Too close to call
DeepSeek-V3.2-Exp: 74.7 (#89), Gemma 4 26B A4B IT: 75.0 (#78)
| Benchmark | DeepSeek-V3.2-Exp | Gemma 4 26B A4B IT |
|---|---|---|
| LMArena Instruction Following | 1413 | 1420 |
Long Context DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 47.6 (#15), Gemma 4 26B A4B IT: 43.6 (#87)
| Benchmark | DeepSeek-V3.2-Exp | Gemma 4 26B A4B IT |
|---|---|---|
| LMArena Longer Query | 1428 | 1427 |
| Fiction.LiveBench | 83.3% | — |
| CL-bench | 13.2% | — |
| CL-bench Life | 9.5% | — |
Writing & Preference DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 62.5 (#73), Gemma 4 26B A4B IT: 58.8 (#110)
| Benchmark | DeepSeek-V3.2-Exp | Gemma 4 26B A4B IT |
|---|---|---|
| LMArena Text | 1425 | 1434 |
| LMArena Creative Writing | 1403 | 1401 |
| EQ-Bench Creative Writing | 1515 | 1305 |
| LMArena Multi-Turn | 1427 | 1442 |
Frequently asked questions
Is DeepSeek-V3.2-Exp better than Gemma 4 26B A4B IT?
DeepSeek-V3.2-Exp is the stronger model overall, scoring 44.9 to 43.1 on the Noometry Index. Gemma 4 26B A4B IT costs 2.4× less per token, which makes it the better buy when DeepSeek-V3.2-Exp's lead doesn't matter for your workload.
Which is cheaper, DeepSeek-V3.2-Exp or Gemma 4 26B A4B IT?
Gemma 4 26B A4B IT is cheaper. It lists at $0.0765 per million input tokens and $0.26 per million output tokens; DeepSeek-V3.2-Exp lists at $0.26 and $0.38.
Is DeepSeek-V3.2-Exp or Gemma 4 26B A4B IT better for coding?
DeepSeek-V3.2-Exp scores higher on coding benchmarks: 46.4 versus 38.8 in the Noometry coding category.
Which has the bigger context window?
Gemma 4 26B A4B IT does, with 262K tokens against 164K.
How many benchmarks do DeepSeek-V3.2-Exp and Gemma 4 26B A4B IT share?
26 benchmarks have published results for both models. DeepSeek-V3.2-Exp has 48 scored results on Noometry and Gemma 4 26B A4B IT has 28.