Model comparison
Gemma 4 26B A4B IT vs GPT-5.6 Terra
GPT-5.6 Terra is the stronger model overall, scoring 59.2 to 43.5 on the Noometry Index. Gemma 4 26B A4B IT costs 42× less per token, which makes it the better buy when GPT-5.6 Terra's lead doesn't matter for your workload.
Last verified . 27 shared benchmarks.
GoogleSummary
- They share 27 benchmarks with published results for both. Gemma 4 26B A4B IT scores higher in 0 categories and GPT-5.6 Terra in 9 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GPT-5.6 Terra leads 60.7 to 21.8.
- The biggest single-benchmark swing is Chess Puzzles: 6% for Gemma 4 26B A4B IT and 54% for GPT-5.6 Terra.
- Gemma 4 26B A4B IT is cheaper at $0.0675 / $0.23 per million input/output tokens, against $2 / $12 for GPT-5.6 Terra.
- GPT-5.6 Terra accepts more context: 1.05M tokens versus 262K.
- Gemma 4 26B A4B IT has downloadable open weights; the other is API-only.
Side by side
| Gemma 4 26B A4B IT | GPT-5.6 Terra | |
|---|---|---|
| Provider | OpenAI | |
| Noometry Index | 43.5 | 59.2 |
| Released | 2026-04-02 | 2026-07-09 |
| Weights | Open | Proprietary |
| Context window | 262K | 1.05M |
| Max output | 33K | 128K |
| Input $ / M tokens | $0.0675 | $2 |
| Output $ / M tokens | $0.23 | $12 |
| Results tracked | 28 | 52 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.6 Terra leads
Gemma 4 26B A4B IT: 39.0 (#164), GPT-5.6 Terra: 57.7 (#19)
| Benchmark | Gemma 4 26B A4B IT | GPT-5.6 Terra |
|---|---|---|
| LMArena WebDev | 1359 | 1522 |
| SciCode | 40% | 55% |
| WeirdML | 35.2% | 78.3% |
| LMArena Coding | 1447 | 1484 |
| ALE-Bench | 927.17 | 1,951 |
| DeepSWE | — | 69.6% |
| FrontierCode | — | 41.3% |
| CursorBench | — | 41.3% |
Agentic & Tool Use Not comparable
Gemma 4 26B A4B IT: —, GPT-5.6 Terra: 40.1 (#25)
| Benchmark | Gemma 4 26B A4B IT | GPT-5.6 Terra |
|---|---|---|
| APEX-Agents | — | 58.2% |
| BALROG | — | 53.2% |
| GDP.pdf | — | 24.7% |
| Vending-Bench 2 | — | 7,343 |
Reasoning GPT-5.6 Terra leads
Gemma 4 26B A4B IT: 21.8 (#213), GPT-5.6 Terra: 60.7 (#21)
| Benchmark | Gemma 4 26B A4B IT | GPT-5.6 Terra |
|---|---|---|
| CritPt | 0% | 30% |
| Chess Puzzles | 6% | 54% |
| LMArena Hard Prompts | 1439 | 1468 |
| DTBench | 74.9% | 93.3% |
| LMCA | 29.7% | 55% |
| Epoch Capabilities Index | 141.85 | 159.62 |
| ARC-AGI-2 | — | 83.9% |
| SimpleBench | — | 48.9% |
| Kagi LLM Benchmark | — | 51.3% |
| NYT Connections (extended) | — | 78.4% |
| ARC-AGI-1 | — | 96.5% |
| Mystery Game Puzzles | — | 35% |
| Surface Evolver Bench | — | 83.8% |
Math GPT-5.6 Terra leads
Gemma 4 26B A4B IT: 47.6 (#67), GPT-5.6 Terra: 81.6 (#12)
| Benchmark | Gemma 4 26B A4B IT | GPT-5.6 Terra |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 82.2% | 99.7% |
| LMArena Math | 1470 | 1466 |
| FrontierMath (Tiers 1-3) | — | 86% |
| FrontierMath Tier 4 | — | 70.7% |
| ProofBench | — | 74% |
Knowledge GPT-5.6 Terra leads
Gemma 4 26B A4B IT: 45.7 (#85), GPT-5.6 Terra: 61.2 (#30)
| Benchmark | Gemma 4 26B A4B IT | GPT-5.6 Terra |
|---|---|---|
| GPQA Diamond | 73.2% | 93.3% |
| LMArena Expert | 1447 | 1492 |
| SimpleQA Verified | — | 43.2% |
| Vectara Hallucination Rate | 5.2% | — |
Multimodal GPT-5.6 Terra leads
Gemma 4 26B A4B IT: 40.6 (#46), GPT-5.6 Terra: 47.3 (#11)
| Benchmark | Gemma 4 26B A4B IT | GPT-5.6 Terra |
|---|---|---|
| LMArena Vision | 1260 | 1271 |
| Blueprint-Bench 2 | — | 30.8% |
| Furniture Assembly | — | 54.2% |
| LMArena Document | — | 1472 |
Multilingual GPT-5.6 Terra leads
Gemma 4 26B A4B IT: 53.1 (#71), GPT-5.6 Terra: 54.4 (#44)
| Benchmark | Gemma 4 26B A4B IT | GPT-5.6 Terra |
|---|---|---|
| LMArena Non-English | 1421 | 1439 |
| LMArena Chinese | 1495 | 1513 |
| LMArena French | 1460 | 1471 |
| LMArena Russian | 1434 | 1450 |
| LMArena Spanish | 1417 | 1448 |
| LMArena German | — | 1460 |
| LMArena Japanese | — | 1457 |
| LMArena Korean | — | 1425 |
Instruction Following GPT-5.6 Terra leads
Gemma 4 26B A4B IT: 74.8 (#82), GPT-5.6 Terra: 76.4 (#40)
| Benchmark | Gemma 4 26B A4B IT | GPT-5.6 Terra |
|---|---|---|
| LMArena Instruction Following | 1420 | 1454 |
Long Context Too close to call
Gemma 4 26B A4B IT: 43.6 (#91), GPT-5.6 Terra: 44.4 (#68)
| Benchmark | Gemma 4 26B A4B IT | GPT-5.6 Terra |
|---|---|---|
| LMArena Longer Query | 1428 | 1451 |
Writing & Preference GPT-5.6 Terra leads
Gemma 4 26B A4B IT: 58.6 (#115), GPT-5.6 Terra: 70.2 (#23)
| Benchmark | Gemma 4 26B A4B IT | GPT-5.6 Terra |
|---|---|---|
| LMArena Text | 1434 | 1447 |
| LMArena Creative Writing | 1402 | 1410 |
| EQ-Bench Creative Writing | 1305 | 1855 |
| LMArena Multi-Turn | 1441 | 1449 |
| EQ-Bench 4 | — | 1234 |
Frequently asked questions
Is Gemma 4 26B A4B IT better than GPT-5.6 Terra?
GPT-5.6 Terra is the stronger model overall, scoring 59.2 to 43.5 on the Noometry Index. Gemma 4 26B A4B IT costs 42× less per token, which makes it the better buy when GPT-5.6 Terra's lead doesn't matter for your workload.
Which is cheaper, Gemma 4 26B A4B IT or GPT-5.6 Terra?
Gemma 4 26B A4B IT is cheaper. It lists at $0.0675 per million input tokens and $0.23 per million output tokens; GPT-5.6 Terra lists at $2 and $12.
Is Gemma 4 26B A4B IT or GPT-5.6 Terra better for coding?
GPT-5.6 Terra scores higher on coding benchmarks: 57.7 versus 39.0 in the Noometry coding category.
Which has the bigger context window?
GPT-5.6 Terra does, with 1.05M tokens against 262K.
How many benchmarks do Gemma 4 26B A4B IT and GPT-5.6 Terra share?
27 benchmarks have published results for both models. Gemma 4 26B A4B IT has 28 scored results on Noometry and GPT-5.6 Terra has 52.