Qwen3.5 122B A10B vs Qwen3.5 9B
Wins 6 of 7 areas
Coding · Agents · Reasoning · Images and charts · Long documents · Following instructions
Wins 0 of 7 areas
—
Qwen3.5 122B A10B is the stronger all-rounder.Qwen3.5 9B is cheaper.
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 software40Qwen3.5 122B A10B4 of 4 tests
- AgentsCarrying out multi-step tasks on its own20Qwen3.5 122B A10B2 of 3 tests · 1 tie
- ReasoningHard problems that need careful thinking20Qwen3.5 122B A10B2 of 3 tests · 1 tie
- Long documentsFinding answers in very long texts20Qwen3.5 122B A10B2 of 2 tests
- Following instructionsDoing exactly what it is asked20Qwen3.5 122B A10B2 of 2 tests
- Images and chartsUnderstanding pictures, charts and video10Qwen3.5 122B A10B1 of 1 test
- FactsGetting facts right instead of making them up11Even1 each
Images and charts 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.
Qwen3.5 9B costs 91% 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 Qwen3.5 122B A10B pulls ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+18.4points ahead
- Recent programming contest problemsLiveCodeBench v6+13.3points ahead
- Very hard expert questions across many subjectsHumanity's Last Exam+10.3points ahead
Where Qwen3.5 9B pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+3.5points 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.
CodingQwen3.5 122B A10B
- Terminal-Bench 2.1Qwen3.5 122B A10B by 18.447.629.2+18.4
- LiveCodeBench v6Qwen3.5 122B A10B by 13.378.965.6+13.3
- SciCodeQwen3.5 122B A10B by 10.239.729.5+10.2
- Terminal-Bench HardQwen3.5 122B A10B by 6.931.124.2+6.9
AgentsQwen3.5 122B A10B
- GDPValQwen3.5 122B A10B by 15.815.80+15.8
- τ-Bench V3 · BankingQwen3.5 122B A10B by 8.315.37+8.3
- Terminal-Bench 4.0tie00.5tie
ReasoningQwen3.5 122B A10B
- Humanity's Last ExamQwen3.5 122B A10B by 10.325.214.9+10.3
- GPQA DiamondQwen3.5 122B A10B by 5.185.780.6+5.1
- CritPttie0.60.3tie
FactsEven
- AA-Omniscience · AccuracyQwen3.5 122B A10B by 824.416.4+8
- AA-Omniscience · Non-hallucinationQwen3.5 9B by 3.512.916.4+3.5
Images and chartsQwen3.5 122B A10B
- MMMU-ProQwen3.5 122B A10B by 5.87569.2+5.8
Long documentsQwen3.5 122B A10B
- AA-LCRQwen3.5 122B A10B by 6.376.370+6.3
- LongBench v2Qwen3.5 122B A10B by 560.255.2+5
Following instructionsQwen3.5 122B A10B
- IFBenchQwen3.5 122B A10B by 975.766.7+9
- Multi-ChallengeQwen3.5 122B A10B by 761.554.5+7
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.
- MLVUQwen3.5 122B A10B by 34.987.352.4+34.9
- Artificial Analysis Coding IndexQwen3.5 122B A10B by 1745.728.7+17
- BFCLv4Qwen3.5 122B A10B by 12.172.260.1+12.1
- AA-OmniscienceQwen3.5 122B A10B by 12-41.5-53.5+12
- SuperGPQAQwen3.5 122B A10B by 8.967.158.2+8.9
- HMMT 2025Qwen3.5 122B A10B by 7.490.382.9+7.4
- IncludeQwen3.5 122B A10B by 7.282.875.6+7.2
- τ²-Bench Telecom (AA run)Qwen3.5 122B A10B by 6.893.686.8+6.8
- DeepPlanningQwen3.5 122B A10B by 6.124.118+6.1
- MMVUQwen3.5 122B A10B by 6.174.768.6+6.1
- MMLU-ProXQwen3.5 122B A10B by 5.982.276.3+5.9
- MMMLUQwen3.5 122B A10B by 5.586.781.2+5.5
- AA IntelligenceQwen3.5 122B A10B by 4.415.611.2+4.4
- MMLU-ProQwen3.5 122B A10B by 4.286.782.5+4.2
- VITA-BenchQwen3.5 122B A10B by 3.833.629.8+3.8
- C-EvalQwen3.5 122B A10B by 3.791.988.2+3.7
- MMLU-ReduxQwen3.5 122B A10B by 2.99491.1+2.9
- AA Agentic IndexQwen3.5 122B A10B by 2.69.67+2.6
- IFEvalQwen3.5 122B A10B by 1.993.491.5+1.9
- t2-benchtie79.579.1tie
Questions people ask
Which is better, Qwen3.5 122B A10B or Qwen3.5 9B?
Qwen3.5 122B A10B wins six of the seven areas where both have results: coding, agents, reasoning, images and charts, long documents and following instructions. Qwen3.5 9B wins none, but costs 91% less. They are level on facts.
Which is better for coding?
Qwen3.5 122B A10B. It wins 4 of the 4 coding tests both models report; Qwen3.5 9B wins none.
Which is cheaper?
Qwen3.5 122B A10B costs $0.40 per million input tokens and $3.20 per million output tokens; Qwen3.5 9B costs $0.14 and $0.20. That makes Qwen3.5 9B about 91% 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 17 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.