Qwen3.7 Plus Preview vs Qwen3.8 Flash Next
Wins 0 of 7 areas
—
Wins 6 of 7 areas
Coding · Agents · Reasoning · Images and charts · Long documents · Following instructions
Qwen3.8 Flash Next is the stronger all-rounder.
Scores updated · 37 tests both models report · How we compare
Where each one wins
Tests won in each of the seven 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.
- CodingWriting and fixing software06Qwen3.8 Flash Next6 of 6 tests
- ReasoningHard problems that need careful thinking03Qwen3.8 Flash Next3 of 3 tests
- AgentsCarrying out multi-step tasks on its own13Qwen3.8 Flash Next3 of 4 tests
- Images and chartsUnderstanding pictures, charts and video01Qwen3.8 Flash Next1 of 2 tests · 1 tie
- Long documentsFinding answers in very long texts01Qwen3.8 Flash Next1 of 1 test
- Following instructionsDoing exactly what it is asked01Qwen3.8 Flash Next1 of 1 test
- FactsGetting facts right instead of making them up11Even1 each
Long documents and following instructions rest on a single test each.
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.
Qwen3.8 Flash Next costs 69% less for the same work.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where Qwen3.8 Flash Next pulls ahead
- Real work tasks from 44 professionsGDPVal+43points ahead
- Customer-service tasks in a simulated bankτ-Bench V3 · Banking+27.9points ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+25.1points ahead
Where Qwen3.7 Plus Preview pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+17.6points ahead
- Completes tasks by operating a computer desktopOSWorld-Verified+17.5points ahead
Every test, side by side
All 37 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingQwen3.8 Flash Next
- Terminal-Bench 2.1Qwen3.8 Flash Next by 25.16186.1+25.1
- SWE-bench MultilingualQwen3.8 Flash Next by 5.275.881+5.2
- SWE-bench ProQwen3.8 Flash Next by 4.957.662.5+4.9
- SciCodeQwen3.8 Flash Next by 4.546.150.6+4.5
- LiveCodeBench v6Qwen3.8 Flash Next by 2.389.691.9+2.3
- SWE-bench VerifiedQwen3.8 Flash Next by 177.778.7+1
AgentsQwen3.8 Flash Next
- GDPValQwen3.8 Flash Next by 4313.456.4+43
- τ-Bench V3 · BankingQwen3.8 Flash Next by 27.917.545.4+27.9
- Terminal-Bench 4.0Qwen3.8 Flash Next by 24.3125.3+24.3
- OSWorld-VerifiedQwen3.7 Plus Preview by 17.573.355.8+17.5
ReasoningQwen3.8 Flash Next
- Humanity's Last ExamQwen3.8 Flash Next by 2.435.638+2.4
- GPQA DiamondQwen3.8 Flash Next by 2.39092.3+2.3
- CritPtQwen3.8 Flash Next by 29.111.1+2
FactsEven
- AA-Omniscience · Non-hallucinationQwen3.7 Plus Preview by 17.672.354.7+17.6
- AA-Omniscience · AccuracyQwen3.8 Flash Next by 222.524.5+2
Images and chartsQwen3.8 Flash Next
- CharXiv (RQ)Qwen3.8 Flash Next by 4.785.990.6+4.7
- MMMU-Protie80.579.8tie
Long documentsQwen3.8 Flash Next
- AA-LCRQwen3.8 Flash Next by 6.77379.7+6.7
Following instructionsQwen3.8 Flash Next
- IFBenchQwen3.8 Flash Next by 3.37881.3+3.3
Other results18 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.
- DeepSWE 1.1Qwen3.8 Flash Next by 42.216.558.7+42.2
- Agents' Last ExamQwen3.8 Flash Next by 3813.251.2+38
- AA Agentic IndexQwen3.8 Flash Next by 36.719.756.4+36.7
- OSWorld 2.0 (partial)Qwen3.8 Flash Next by 30.821.552.3+30.8
- JobBenchQwen3.8 Flash Next by 28.127.655.7+28.1
- BabyVisionQwen3.7 Plus Preview by 23.670.446.8+23.6
- Toolathlon VerifiedQwen3.8 Flash Next by 22.950.673.5+22.9
- Artificial Analysis Coding IndexQwen3.8 Flash Next by 17.255.973.1+17.2
- AA IntelligenceQwen3.8 Flash Next by 14.625.239.8+14.6
- AA-OmniscienceQwen3.7 Plus Preview by 10.81.1-9.7+10.8
- NL2RepoQwen3.8 Flash Next by 741.148.1+7
- MATH-VisionQwen3.8 Flash Next by 5.490.395.7+5.4
- AndroidWorldQwen3.8 Flash Next by 3.58184.5+3.5
- ERQAQwen3.8 Flash Next by 2.569.872.3+2.5
- MMLU-ProXQwen3.7 Plus Preview by 2.485.483+2.4
- RealWorldQAQwen3.8 Flash Next by 1.686.988.5+1.6
- WMT24++Qwen3.8 Flash Next by 1.684.686.2+1.6
- LVBenchtie76.276.6tie
Questions people ask
Which is better, Qwen3.7 Plus Preview or Qwen3.8 Flash Next?
Qwen3.8 Flash Next wins six of the seven areas where both have results: coding, agents, reasoning, images and charts, long documents and following instructions. Qwen3.7 Plus Preview wins none. They are level on facts.
Which is better for coding?
Qwen3.8 Flash Next. It wins 6 of the 6 coding tests both models report; Qwen3.7 Plus Preview wins none.
Which is cheaper?
Qwen3.7 Plus Preview costs $0.40 per million input tokens and $1.60 per million output tokens; Qwen3.8 Flash Next costs $0.15 and $0.47. That makes Qwen3.8 Flash Next about 69% cheaper for the same work.
How do you compare the two?
We use the 37 benchmark tests both models have published scores on. The verdict counts the 19 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 18 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.