Qwen3.8 27B wins more areas, narrowly.
Scores updated · 8 tests both models report · How we compare
Where each one wins
Tests won in each of the two areas where both have results. Each piece is one test, so a longer bar means more evidence; grey means the two scored within a point of each other.
Images and charts rests on a single test.
What it costs
Prices per million tokens, roughly 750,000 words. The bars show the cost of a million tokens read plus a million written.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where Qwen3.8 27B pulls ahead
- Harder college exam questions with imagesMMMU-Pro+3.1points ahead
- Fixes real GitHub issues in many programming languagesSWE-bench Multilingual+1.8points ahead
Where TML-d24-varloop-research pulls ahead
- Fixes real GitHub issues in Python projectsSWE-bench Verified+1.8points ahead
Every test, side by side
All 8 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingEven
- SWE-bench MultilingualQwen3.8 27B by 1.873.872+1.8
- SWE-bench VerifiedTML-d24-varloop-research by 1.877.879.6+1.8
Other results5 tests, not counted
Tests outside the eight areas. They are not counted above: several are summary scores built from other tests, or the same test under another name.
- JobBenchTML-d24-varloop-research by 29.933.463.3+29.9
- DeepSWE 1.1TML-d24-varloop-research by 642.248.2+6
- SWEBench Pro v2 (Hard)TML-d24-varloop-research by 4.673.878.4+4.6
- GDP (Surge AI)TML-d24-varloop-research by 3.616.420+3.6
- OmniDocBench v1.6TML-d24-varloop-research by 191.592.5+1
Questions people ask
Which is better, Qwen3.8 27B or TML-d24-varloop-research?
Qwen3.8 27B wins one of the two areas where both have results: images and charts. TML-d24-varloop-research wins none. They are level on coding.
Which is better for coding?
Neither. They win 1 coding test each of the 2 both models report.
How do you compare the two?
We use the 8 benchmark tests both models have published scores on. The verdict counts the 3 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 5 are listed but not counted, because several are summary scores or repeat a test. Each score is the one shown on the model's own page.