GPT-5 vs Qwen3 Next 80B A3B
Wins 6 of 6 areas
Coding · Agents · Reasoning · Facts · Long documents · Following instructions
Wins 0 of 6 areas
—
GPT-5 is the stronger all-rounder.Qwen3 Next 80B A3B is cheaper.
Scores updated · 34 tests both models report · How we compare
Where each one wins
Tests won in each of the six 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 software40GPT-54 of 4 tests
- ReasoningHard problems that need careful thinking30GPT-53 of 3 tests
- AgentsCarrying out multi-step tasks on its own20GPT-52 of 2 tests
- FactsGetting facts right instead of making them up21GPT-52 of 3 tests
- Long documentsFinding answers in very long texts10GPT-51 of 1 test
- Following instructionsDoing exactly what it is asked10GPT-51 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 Next 80B A3B 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 pulls ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+28.5points ahead
- Hard command-line tasks in a real terminalTerminal-Bench Hard+22.8points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+21.3points ahead
Where Qwen3 Next 80B A3B pulls ahead
No clear win on a test scored out of 100.
Every test, side by side
All 34 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGPT-5
- Terminal-Bench 2.1GPT-5 by 28.535.26.7+28.5
- Terminal-Bench HardGPT-5 by 22.832.69.8+22.8
- LiveCodeBench v6GPT-5 by 18.38768.7+18.3
- SciCodeGPT-5 by 4.142.938.8+4.1
ReasoningGPT-5
- Humanity's Last ExamGPT-5 by 15.928.512.6+15.9
- GPQA DiamondGPT-5 by 9.585.475.9+9.5
- CritPtGPT-5 by 5.75.70+5.7
FactsGPT-5
- AA-Omniscience · AccuracyGPT-5 by 21.340.319+21.3
- Vectara HHEM hallucination ratelower is betterQwen3 Next 80B A3B by 5.815.19.3+5.8
- AA-Omniscience · Non-hallucinationGPT-5 by 4.817.813+4.8
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.
- τ²-Bench Telecom (AA run)GPT-5 by 43.384.841.5+43.3
- AA-OmniscienceGPT-5 by 42.7-8.7-51.4+42.7
- AA Agentic IndexGPT-5 by 24.426.52.1+24.4
- Artificial Analysis Coding IndexGPT-5 by 20.437.817.4+20.4
- HMMT 2025GPT-5 by 19.493.373.9+19.4
- τ²-Bench (Retail)GPT-5 by 13.381.167.8+13.3
- HarmBenchGPT-5 by 1397.684.6+13
- AA IntelligenceGPT-5 by 11.82311.2+11.8
- vectara_avg_summary_lengthGPT-5 by 91.8 rating points162.770.9+91.8 rating
- AIME 2025GPT-5 by 6.894.687.8+6.8
- vectara_factual_consistencyQwen3 Next 80B A3B by 5.884.990.7+5.8
- vectara_answer_rateGPT-5 by 5.599.994.4+5.5
- bbqGPT-5 by 5.196.891.7+5.1
- MMLU-ProGPT-5 by 4.487.182.7+4.4
- MMLU-ReduxGPT-5 by 2.895.392.5+2.8
- Tau2 airlineGPT-5 by 2.162.660.5+2.1
- XSTestQwen3 Next 80B A3B by 1.297.198.3+1.2
- AIR-Bench 2024GPT-5 by 187.786.7+1
- anthropic_red_teamtie99.199.9tie
- simple_safety_teststie99.899.5tie
Questions people ask
Which is better, GPT-5 or Qwen3 Next 80B A3B?
GPT-5 wins all six areas where both have results: coding, agents, reasoning, facts, long documents and following instructions. Qwen3 Next 80B A3B wins none, but costs 88% less.
Which is better for coding?
GPT-5. It wins 4 of the 4 coding tests both models report; Qwen3 Next 80B A3B wins none.
Which is cheaper?
GPT-5 costs $1.25 per million input tokens and $10.00 per million output tokens; Qwen3 Next 80B A3B costs $0.15 and $1.20. That makes Qwen3 Next 80B A3B about 88% cheaper for the same work.
How do you compare the two?
We use the 34 benchmark tests both models have published scores on. The verdict counts the 14 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.