GPT-5.5 vs Qwen3.7 Plus Preview
Wins 5 of 8 areas
Coding · Agents · Reasoning · Math · Long documents
Wins 1 of 8 areas
Following instructions
GPT-5.5 is the stronger all-rounder.Qwen3.7 Plus Preview is cheaper and better at following instructions.
Scores updated · 41 tests both models report · How we compare
Where each one wins
Tests won in each of the eight areas we test. Each piece is one test, so a longer bar means more evidence; grey means the two scored within a point of each other.
- AgentsCarrying out multi-step tasks on its own60GPT-5.56 of 6 tests
- CodingWriting and fixing software40GPT-5.54 of 4 tests
- ReasoningHard problems that need careful thinking30GPT-5.53 of 3 tests
- MathCompetition and research-level math20GPT-5.52 of 2 tests
- Long documentsFinding answers in very long texts10GPT-5.51 of 1 test
- Following instructionsDoing exactly what it is asked01Qwen3.7 Plus Preview1 of 1 test
- FactsGetting facts right instead of making them up11Even1 each
- Images and chartsUnderstanding pictures, charts and video11Even1 each · 1 tie
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.
Qwen3.7 Plus Preview costs 94% less for the same work.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where GPT-5.5 pulls ahead
- Unpublished advanced math problemsFrontierMath Tiers 1-3 (v2)+50.9points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+35.5points ahead
- Real work tasks from 44 professionsGDPVal+29.1points ahead
Where Qwen3.7 Plus Preview pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+61.3points ahead
- Reads and reasons about text in imagesOCRBenchv2+6points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+2.1points ahead
Every test, side by side
All 41 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGPT-5.5
- Terminal-Bench 2.1GPT-5.5 by 23.384.361+23.3
- Terminal-Bench HardGPT-5.5 by 13.660.647+13.6
- SciCodeGPT-5.5 by 9.755.846.1+9.7
- SWE-bench ProGPT-5.5 by 158.657.6+1
AgentsGPT-5.5
- GDPValGPT-5.5 by 29.142.513.4+29.1
- τ-Bench V3 · BankingGPT-5.5 by 21.53917.5+21.5
- AA ApexAgentsGPT-5.5 by 15.337.722.4+15.3
- Terminal-Bench 4.0GPT-5.5 by 13.614.61+13.6
- OSWorld-VerifiedGPT-5.5 by 5.478.773.3+5.4
- MCP AtlasGPT-5.5 by 2.175.373.2+2.1
ReasoningGPT-5.5
- CritPtGPT-5.5 by 1827.19.1+18
- Humanity's Last ExamGPT-5.5 by 10.245.835.6+10.2
- GPQA DiamondGPT-5.5 by 3.593.590+3.5
FactsEven
- AA-Omniscience · Non-hallucinationQwen3.7 Plus Preview by 61.31172.3+61.3
- AA-Omniscience · AccuracyGPT-5.5 by 35.55822.5+35.5
Images and chartsEven
- OCRBenchv2Qwen3.7 Plus Preview by 661.167.1+6
- LMArena · VisionGPT-5.5 by 18 rating points12951277+18 rating
- MMMU-Protie79.980.5tie
MathGPT-5.5
- FrontierMath Tiers 1-3 (v2)GPT-5.5 by 50.985.334.4+50.9
- HMMT Feb. 2026GPT-5.5 by 5.698.592.9+5.6
Following instructionsQwen3.7 Plus Preview
- IFBenchQwen3.7 Plus Preview by 2.175.978+2.1
Other results19 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 1.1GPT-5.5 by 50.56716.5+50.5
- WorldVQAQwen3.7 Plus Preview by 26.534.661.1+26.5
- SimpleVQAQwen3.7 Plus Preview by 23.158.681.7+23.1
- Toolathlon VerifiedGPT-5.5 by 22.973.550.6+22.9
- AA-OmniscienceGPT-5.5 by 19.420.51.1+19.4
- Artificial Analysis Coding IndexGPT-5.5 by 1974.955.9+19
- AA Agentic IndexGPT-5.5 by 17.637.319.7+17.6
- BabyVisionQwen3.7 Plus Preview by 14.555.970.4+14.5
- Finance Agent v2GPT-5.5 by 13.651.838.2+13.6
- Agents' Last ExamGPT-5.5 by 13.426.613.2+13.4
- AA IntelligenceGPT-5.5 by 13.238.425.2+13.2
- GDP (Surge AI)GPT-5.5 by 12.724.912.2+12.7
- Terminal-Bench 2.0GPT-5.5 by 12.482.770.3+12.4
- JobBenchGPT-5.5 by 10.738.327.6+10.7
- NL2RepoGPT-5.5 by 9.650.741.1+9.6
- ERQAQwen3.7 Plus Preview by 5.364.569.8+5.3
- RealWorldQAQwen3.7 Plus Preview by 4.782.286.9+4.7
- SuperGPQAGPT-5.5 by 1.372.771.4+1.3
- τ²-Bench Telecom (AA run)tie93.993tie
Questions people ask
Which is better, GPT-5.5 or Qwen3.7 Plus Preview?
GPT-5.5 wins five of the eight areas we test: coding, agents, reasoning, math and long documents. Qwen3.7 Plus Preview wins following instructions, and costs 94% less. They are level on facts and images and charts.
Which is better for coding?
GPT-5.5. It wins 4 of the 4 coding tests both models report; Qwen3.7 Plus Preview wins none.
Which is cheaper?
GPT-5.5 costs $5.00 per million input tokens and $30.00 per million output tokens; Qwen3.7 Plus Preview costs $0.40 and $1.60. That makes Qwen3.7 Plus Preview about 94% cheaper for the same work.
How do you compare the two?
We use the 41 benchmark tests both models have published scores on. The verdict counts the 22 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 19 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.