Qwen3.5-4B-Base
Qwen3.5-4B-Base.
8
Benchmarks with a score
0
Ranked first, out of five or more models
0
In the top three
Qwen3.5-4B-Base
Open weights Small
No Epoch Capabilities Index score yet
- Angry Tweets 6.93% #143 /149
- QIMMA 52.3% #21 /44
- Global PIQA 59.5% #40 /59
How does it do against its rivals?
Qwen3.5-4B-Base beats Gemma 4 E2B on 5 of the 8 benchmarks they share. Gemma 4 31B IT does best against it, ahead on 8 of 8.
-
Gemma 4 E2B 5 ahead, 3 behind Compare
-
Gemma 4 E4B 1 ahead, 7 behind Compare
-
Gemma 4 31B IT 0 ahead, 8 behind Compare
Sources: Official leaderboards, Epoch AI (CC BY 4.0) and labs’ model cards
Where it ranks
Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.
Knowledge and language
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| QIMMA | 52.3% | #21 of 44 | Official leaderboard | |
| Uncheatable Eval | 7.28 | #16 of 33 | Official leaderboard | |
| FoSent | 45.5% | #72 of 147 | Official leaderboard | |
| Global PIQA | 59.5% | #40 of 59 | Official leaderboard | |
| EuroEval | 3.00 | #61 of 75 | Official leaderboard | |
| EuroEval ERRNews | 21.1 | #67 of 76 | Official leaderboard | |
| Angry Tweets | 6.93% | #143 of 149 | Official leaderboard |
Safety and security
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| nocot-bench (NCRI) | 46.0 | #183 of 201 | Official leaderboard |
Who measured it?
Independent runs cover all 8 of its benchmarks.
- An independent run only 8
Sources: Official leaderboards, Epoch AI (CC BY 4.0) and labs’ model cards