Rankings / Google
Gemma 4 26B A4b
open weightsreleased 2026-04-02
BenchAtlas Index
as of 2026-09-11
thinking
rank #279
7 families · 4 categories · medium
no reasoning
rank #324
7 families · 4 categories · medium
Benchmark evidence
34 results
agentic coding
| τ²-Bench | thinking | 43.6 % | independent | Artificial Analysis |
| τ²-Bench | no reasoning | 40.4 % | independent | Artificial Analysis |
| τ²-Bench subset=banking | thinking | 12.0 % | independent | Artificial Analysis |
coding
| SciCode | thinking | 40.0 % | independent | Artificial Analysis |
| SciCode | no reasoning | 37.3 % | independent | Artificial Analysis |
external indices
| AA Coding Index | thinking | 39.3 points | independent | Artificial Analysis |
| Intelligence Index v4.1 | thinking | 16.7 points | independent | Artificial Analysis |
| Intelligence Index v4.1 | no reasoning | 13.1 points | independent | Artificial Analysis |
| ECI | 141.9 points | independent | Epoch AI Benchmarking Hub |
human preference
| Coding (style control)older version arena=text · category=coding · style_control=true | 1481.0 1465.8–1496.2 | community | LMArena Leaderboard Dataset | |
| Industry Software And It Services (style control)older version arena=text · category=industry_software_and_it_services · style_control=true | 1475.2 1463.0–1487.5 | community | LMArena Leaderboard Dataset | |
| Hard Prompts English (style control)older version arena=text · category=hard_prompts_english · style_control=true | 1473.2 1456.8–1489.5 | community | LMArena Leaderboard Dataset | |
| Hard Prompts (style control)older version arena=text · category=hard_prompts · style_control=true | 1461.8 1451.8–1471.9 | community | LMArena Leaderboard Dataset | |
| Industry Life And Physical And Social Science (style control)older version arena=text · category=industry_life_and_physical_and_social_science · style_control=true | 1452.6 1433.4–1471.8 | community | LMArena Leaderboard Dataset | |
| English (style control)older version arena=text · category=english · style_control=true | 1450.9 1438.6–1463.1 | community | LMArena Leaderboard Dataset | |
| Longer Query (style control)older version arena=text · category=longer_query · style_control=true | 1448.8 1434.5–1463.1 | community | LMArena Leaderboard Dataset | |
| Industry Business And Management And Financial Operations (style control)older version arena=text · category=industry_business_and_management_and_financial_operations · style_control=true | 1448.1 1430.6–1465.6 | community | LMArena Leaderboard Dataset | |
| Multi Turn (style control)older version arena=text · category=multi_turn · style_control=true | 1447.1 1429.6–1464.5 | community | LMArena Leaderboard Dataset | |
| Russian (style control)older version arena=text · category=russian · style_control=true | 1441.3 1422.2–1460.3 | community | LMArena Leaderboard Dataset | |
| Instruction Following (style control)older version arena=text · category=instruction_following · style_control=true | 1438.8 1424.8–1452.7 | community | LMArena Leaderboard Dataset | |
| Overall (style control) arena=text · category=overall · style_control=true | 1438.5 1430.8–1446.1 | community | LMArena Leaderboard Dataset | |
| Exclude Ties (style control)older version arena=text · category=exclude_ties · style_control=true | 1435.5 1424.3–1446.6 | community | LMArena Leaderboard Dataset | |
| Non English (style control)older version arena=text · category=non_english · style_control=true | 1423.7 1414.1–1433.2 | community | LMArena Leaderboard Dataset | |
| Industry Entertainment And Sports And Media (style control)older version arena=text · category=industry_entertainment_and_sports_and_media · style_control=true | 1410.0 1392.0–1428.1 | community | LMArena Leaderboard Dataset | |
| Industry Writing And Literature And Language (style control)older version arena=text · category=industry_writing_and_literature_and_language · style_control=true | 1409.9 1394.3–1425.6 | community | LMArena Leaderboard Dataset | |
| Creative Writing (style control)older version arena=text · category=creative_writing · style_control=true | 1403.8 1384.9–1422.7 | community | LMArena Leaderboard Dataset |
knowledge science
| GPQA Diamond implementation=artificial-analysis | thinking | 79.2 % | independent | Artificial Analysis |
| GPQA Diamond implementation=artificial-analysis | no reasoning | 71.4 % | independent | Artificial Analysis |
| Humanity's Last Exam implementation=artificial-analysis | thinking | 19.3 % | independent | Artificial Analysis |
| Humanity's Last Exam implementation=artificial-analysis | no reasoning | 11.5 % | independent | Artificial Analysis |
long context instruction
| AA-LCR | thinking | 65.7 % | independent | Artificial Analysis |
| AA-LCR | no reasoning | 42.3 % | independent | Artificial Analysis |
| IFBench | thinking | 72.5 % | independent | Artificial Analysis |
| IFBench | no reasoning | 45.4 % | independent | Artificial Analysis |
Agent + model results
systems, not bare-model scores
| agent + model Artificial Analysis harness + Gemma 4 26B A4b | Terminal-Bench 2.1 | 39.0 % | independent | Artificial Analysis |
| agent + model Artificial Analysis harness + Gemma 4 26B A4b | Terminal-Bench Hard | 25.0 % | independent | Artificial Analysis |
| agent + model Artificial Analysis harness + Gemma 4 26B A4b | Terminal-Bench Hard | 13.6 % | independent | Artificial Analysis |
These scores measure the whole agent system (scaffold, tools, budgets) — they are never merged into the bare model’s numbers.