Apple: diffucoder-7b-base
7 benchmarks
Leaderboard position
#820
Overall rank (of 1121 models)
425
Elo rating
7
Benchmarks
Top 73.1%
Percentile
In Apple (of 13 models)
Above: Qwen: qwen3-omni #819
Below: Anthropic: claude-3-haiku #821
Benchmark highlights
Best benchmark
MBPP
74.2
Worst benchmark
BigCodeBench Hard
12.8
Nearest competitor from another company
The model from another company with the most similar benchmark scores.
appeared as nearest in 1 of 7 benchmarks
1/7
14.3%
1/7
14.3%
1/7
14.3%
1/7
14.3%
1/7
14.3%
| Model | Votes | % of benchmarks |
|---|---|---|
| Yandex: gpt-5-lite-pretrain | 1 | 14.3% |
| DeepSeek: deepseek-r1-distill-qwen-1.5b | 1 | 14.3% |
| Inception: mercury | 1 | 14.3% |
| Technology Innovation Institute: falcon-h1-1.5b-deep-base | 1 | 14.3% |
| Qwen: qwen3-1.7b-base | 1 | 14.3% |
| Qwen: qwen2.5-coder-7b-instruct | 1 | 14.3% |
Benchmarks
Predicted scores
Estimated from 25 similar models and the global Elo rating. These are projections, not measured results.