GPT-5.5 vs Qwen3.5 397B A17B
Wins 7 of 8 areas
Coding · Agents · Reasoning · Facts · Images and charts · Math · Long documents
Wins 1 of 8 areas
Following instructions
GPT-5.5 is the stronger all-rounder.Qwen3.5 397B A17B is cheaper and better at following instructions.
Scores updated · 38 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 software50GPT-5.55 of 5 tests
- MathCompetition and research-level math40GPT-5.54 of 4 tests
- ReasoningHard problems that need careful thinking30GPT-5.53 of 3 tests
- FactsGetting facts right instead of making them up20GPT-5.52 of 3 tests · 1 tie
- Images and chartsUnderstanding pictures, charts and video20GPT-5.52 of 2 tests
- Long documentsFinding answers in very long texts10GPT-5.51 of 1 test
- Following instructionsDoing exactly what it is asked01Qwen3.5 397B A17B1 of 1 test
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.
Qwen3.5 397B A17B costs 88% 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 GPT-5.5 pulls ahead
- Proof problems from the US Math OlympiadUSAMO 2026+61.9points ahead
- Unpublished advanced math problemsFrontierMath Tiers 1-3 (v2)+54.1points ahead
- Short factual questions, answered correctlySimpleQA Verified+37points ahead
Where Qwen3.5 397B A17B pulls ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+2.9points ahead
Every test, side by side
All 38 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 3384.351.3+33
- Terminal-Bench HardGPT-5.5 by 19.760.640.9+19.7
- LMArena · WebDevGPT-5.5 by 113 rating points15131400+113 rating
- SciCodeGPT-5.5 by 1155.844.8+11
- SWE-bench ProGPT-5.5 by 7.758.650.9+7.7
AgentsGPT-5.5
- GDPValGPT-5.5 by 27.942.714.8+27.9
- τ-Bench V3 · BankingGPT-5.5 by 25.63913.4+25.6
- AA ApexAgentsGPT-5.5 by 22.437.715.3+22.4
- BrowseCompGPT-5.5 by 15.484.469+15.4
- Terminal-Bench 4.0GPT-5.5 by 14.614.60+14.6
- AA IT-Bench SREGPT-5.5 by 11.745.834.1+11.7
ReasoningGPT-5.5
- CritPtGPT-5.5 by 25.427.11.7+25.4
- Humanity's Last ExamGPT-5.5 by 16.845.829+16.8
- GPQA DiamondGPT-5.5 by 4.293.589.3+4.2
FactsGPT-5.5
- SimpleQA VerifiedGPT-5.5 by 376326+37
- AA-Omniscience · AccuracyGPT-5.5 by 27.25830.8+27.2
- AA-Omniscience · Non-hallucinationtie1111.1tie
Images and chartsGPT-5.5
- LMArena · VisionGPT-5.5 by 28 rating points12921264+28 rating
- MMMU-ProGPT-5.5 by 2.679.977.3+2.6
MathGPT-5.5
- USAMO 2026GPT-5.5 by 61.998.236.3+61.9
- FrontierMath Tiers 1-3 (v2)GPT-5.5 by 54.185.331.2+54.1
- HMMT Feb. 2026GPT-5.5 by 10.698.587.9+10.6
- AIME 2026GPT-5.5 by 8.710091.3+8.7
Following instructionsQwen3.5 397B A17B
- IFBenchQwen3.5 397B A17B by 2.975.978.8+2.9
Other results13 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.
- GDPval-AA v2GPT-5.5 by 547 rating points1509962+547 rating
- AA-OmniscienceGPT-5.5 by 51.220.5-30.8+51.2
- Toolathlon VerifiedGPT-5.5 by 32.873.540.7+32.8
- Terminal-Bench 2.0GPT-5.5 by 30.282.752.5+30.2
- AA Agentic IndexGPT-5.5 by 26.737.310.6+26.7
- Artificial Analysis Coding IndexGPT-5.5 by 26.774.948.2+26.7
- AA IntelligenceGPT-5.5 by 2038.418.4+20
- NL2RepoGPT-5.5 by 18.550.732.2+18.5
- ToolathlonGPT-5.5 by 17.355.638.3+17.3
- HLE (with tools)GPT-5.5 by 3.952.248.3+3.9
- HMMT Nov. 2025GPT-5.5 by 3.896.592.7+3.8
- SuperGPQAGPT-5.5 by 2.372.770.4+2.3
- τ²-Bench Telecom (AA run)Qwen3.5 397B A17B by 1.793.995.6+1.7
Questions people ask
Which is better, GPT-5.5 or Qwen3.5 397B A17B?
GPT-5.5 wins seven of the eight areas we test: coding, agents, reasoning, facts, images and charts, math and long documents. Qwen3.5 397B A17B wins following instructions, and costs 88% less.
Which is better for coding?
GPT-5.5. It wins 5 of the 5 coding tests both models report; Qwen3.5 397B A17B wins none.
Which is cheaper?
GPT-5.5 costs $5.00 per million input tokens and $30.00 per million output tokens; Qwen3.5 397B A17B costs $0.60 and $3.60. That makes Qwen3.5 397B A17B about 88% cheaper for the same work.
How do you compare the two?
We use the 38 benchmark tests both models have published scores on. The verdict counts the 25 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 13 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.