Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Long-context understanding on LongBench V1 (Context Length Buckets)
Loading...
50.48
Accuracy (0–4k Context)
Sharer-only
29.7216
35.1108
40.5
45.8892
Oct 3, 2025
Accuracy (0–4k Context)
Accuracy (4–8k Context)
Accuracy (8k+ Context)
Average Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy (0–4k Context)
Accuracy (4–8k Context)
Accuracy (8k+ Context)
Average Accuracy
Sharer-only
Sharer=Qwen3-4B, Recei...
2025.10
50.48
48.28
44.36
47.9
C2C
Sharer=Qwen3-4B, Recei...
2025.10
41.46
37.57
34.23
37.97
T2T
Sharer=Qwen3-4B, Recei...
2025.10
38.52
35.79
33.79
36.18
Receiver-only
Sharer=Qwen3-4B, Recei...
2025.10
30.52
26.03
25.99
27.63
Feedback
Search any
task
Search any
task