GPT-5.5 vs Qwen3.8 Max Preview
Wins 4 of 8 areas
Reasoning · Facts · Math · Long documents
Wins 4 of 8 areas
Coding · Agents · Images and charts · Following instructions
The two are evenly matched.GPT-5.5 is better at reasoning and math; Qwen3.8 Max Preview at agents and coding.
Both rank among the ten best models we track in agents and long documents.
Scores updated · 47 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 own15Qwen3.8 Max Preview5 of 6 tests
- CodingWriting and fixing software24Qwen3.8 Max Preview4 of 6 tests
- Images and chartsUnderstanding pictures, charts and video02Qwen3.8 Max Preview2 of 2 tests
- Following instructionsDoing exactly what it is asked02Qwen3.8 Max Preview2 of 2 tests
- ReasoningHard problems that need careful thinking30GPT-5.53 of 4 tests · 1 tie
- MathCompetition and research-level math30GPT-5.53 of 3 tests
- FactsGetting facts right instead of making them up21GPT-5.52 of 3 tests
- Long documentsFinding answers in very long texts10GPT-5.51 of 1 test
Long documents 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.
Qwen3.8 Max Preview costs 77% less for the same work.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where Qwen3.8 Max Preview pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+60.2points ahead
- Complex command-line tasks across many fieldsTerminal-Bench 4.0+24.3points ahead
- Real work tasks from 44 professionsGDPVal+15.8points ahead
Where GPT-5.5 pulls ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+26.3points ahead
- Extremely hard research-level math problemsFrontierMath Tier 4+26.2points ahead
- Short factual questions, answered correctlySimpleQA Verified+15.7points ahead
Every test, side by side
All 47 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingQwen3.8 Max Preview
- LMArena · WebDevQwen3.8 Max Preview by 185 rating points14871672+185 rating
- LiveBench · Agentic CodingQwen3.8 Max Preview by 10.75464.7+10.7
- LiveBench · CodingGPT-5.5 by 9.382.272.9+9.3
- SWE-bench ProQwen3.8 Max Preview by 9.158.667.7+9.1
- Terminal-Bench 2.1Qwen3.8 Max Preview by 4.584.388.8+4.5
- SciCodeGPT-5.5 by 3.755.852.1+3.7
AgentsQwen3.8 Max Preview
- Terminal-Bench 4.0Qwen3.8 Max Preview by 24.314.638.9+24.3
- GDPValQwen3.8 Max Preview by 15.842.558.3+15.8
- τ-Bench V3 · BankingQwen3.8 Max Preview by 8.83947.8+8.8
- OSWorld-VerifiedQwen3.8 Max Preview by 7.478.786.1+7.4
- AA IT-Bench SREGPT-5.5 by 5.545.840.3+5.5
- AA ApexAgentsQwen3.8 Max Preview by 4.737.742.4+4.7
ReasoningGPT-5.5
- CritPtGPT-5.5 by 9.427.117.7+9.4
- Humanity's Last ExamGPT-5.5 by 2.745.843.1+2.7
- LiveBench · ReasoningGPT-5.5 by 1.589.788.2+1.5
- GPQA Diamondtie93.592.8tie
FactsGPT-5.5
- AA-Omniscience · Non-hallucinationQwen3.8 Max Preview by 60.21171.2+60.2
- AA-Omniscience · AccuracyGPT-5.5 by 26.35831.7+26.3
- SimpleQA VerifiedGPT-5.5 by 15.76347.3+15.7
Images and chartsQwen3.8 Max Preview
- MMMU-ProQwen3.8 Max Preview by 2.979.982.8+2.9
- LMArena · VisionQwen3.8 Max Preview by 22 rating points12921314+22 rating
MathGPT-5.5
- FrontierMath Tier 4GPT-5.5 by 26.272.546.3+26.2
- FrontierMath Tiers 1-3 (v2)GPT-5.5 by 10.685.374.7+10.6
- LiveBench · MathematicsGPT-5.5 by 4.695.991.3+4.6
Following instructionsQwen3.8 Max Preview
- IFBenchQwen3.8 Max Preview by 6.975.982.8+6.9
- LiveBench · Instruction FollowingQwen3.8 Max Preview by 3.470.774.1+3.4
Other results20 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.
- Agents' Last ExamQwen3.8 Max Preview by 25.826.652.4+25.8
- GDPval-AA v2Qwen3.8 Max Preview by 230 rating points15091739+230 rating
- ToolathlonQwen3.8 Max Preview by 16.955.672.5+16.9
- JobBenchQwen3.8 Max Preview by 15.138.353.4+15.1
- DeepSWEGPT-5.5 by 13.47056.6+13.4
- ERQAQwen3.8 Max Preview by 13.364.577.8+13.3
- AA Agentic IndexQwen3.8 Max Preview by 12.337.349.6+12.3
- DeepSWE 1.1GPT-5.5 by 10.46756.6+10.4
- AA-OmniscienceGPT-5.5 by 8.520.512+8.5
- livebench_languageGPT-5.5 by 7.787.479.7+7.7
- AA IntelligenceQwen3.8 Max Preview by 738.445.4+7
- RealWorldQAQwen3.8 Max Preview by 5.882.288+5.8
- MLS Bench Litelower is betterGPT-5.5 by 5.535.541+5.5
- NL2RepoQwen3.8 Max Preview by 5.250.755.9+5.2
- HLE (with tools)Qwen3.8 Max Preview by 452.256.2+4
- HealthBenchQwen3.8 Max Preview by 3.756.560.2+3.7
- CyberGymGPT-5.5 by 3.381.878.5+3.3
- livebench_data_analysisGPT-5.5 by 3.281.678.4+3.2
- Artificial Analysis Coding IndexGPT-5.5 by 3.174.971.8+3.1
- Toolathlon VerifiedGPT-5.5 by 173.572.5+1
Questions people ask
Which is better, GPT-5.5 or Qwen3.8 Max Preview?
GPT-5.5 and Qwen3.8 Max Preview each win four of the eight areas we test. GPT-5.5 wins reasoning, facts, math and long documents; Qwen3.8 Max Preview wins coding, agents, images and charts and following instructions. Qwen3.8 Max Preview costs 77% less.
Which is better for coding?
Qwen3.8 Max Preview. It wins 4 of the 6 coding tests both models report; GPT-5.5 wins 2.
Which is cheaper?
GPT-5.5 costs $5.00 per million input tokens and $30.00 per million output tokens; Qwen3.8 Max Preview costs $2.00 and $6.00. That makes Qwen3.8 Max Preview about 77% cheaper for the same work.
How do you compare the two?
We use the 47 benchmark tests both models have published scores on. The verdict counts the 27 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 20 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.