Nemotron 3 Ultra 550B A55B vs Qwen3.5 122B A10B
Wins 4 of 6 areas
Coding · Reasoning · Long documents · Following instructions
Wins 1 of 6 areas
Agents
Nemotron 3 Ultra 550B A55B is the stronger all-rounder.Qwen3.5 122B A10B is better at agents.
Scores updated · 39 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 software51Nemotron 3 Ultra 550B A55B5 of 7 tests · 1 tie
- ReasoningHard problems that need careful thinking30Nemotron 3 Ultra 550B A55B3 of 3 tests
- Long documentsFinding answers in very long texts20Nemotron 3 Ultra 550B A55B2 of 2 tests
- Following instructionsDoing exactly what it is asked20Nemotron 3 Ultra 550B A55B2 of 2 tests
- AgentsCarrying out multi-step tasks on its own12Qwen3.5 122B A10B2 of 4 tests · 1 tie
- FactsGetting facts right instead of making them up11Even1 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.
Nemotron 3 Ultra 550B A55B costs 14% less for the same work.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where Nemotron 3 Ultra 550B A55B pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+57.4points ahead
- Recent programming contest problemsLiveCodeBench v6+10.1points ahead
- Real work tasks from 44 professionsGDPVal+10points ahead
Where Qwen3.5 122B A10B pulls ahead
- Finds hard-to-locate facts by browsing the webBrowseComp+19.4points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+1.8points ahead
- Fixes real GitHub issues in Python projectsSWE-bench Verified+1.3points ahead
Every test, side by side
All 39 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingNemotron 3 Ultra 550B A55B
- LiveCodeBench v6Nemotron 3 Ultra 550B A55B by 10.18978.9+10.1
- Terminal-Bench 2.1Nemotron 3 Ultra 550B A55B by 6.353.947.6+6.3
- SWE-bench MultilingualNemotron 3 Ultra 550B A55B by 5.767.762+5.7
- Terminal-Bench HardNemotron 3 Ultra 550B A55B by 5.336.431.1+5.3
- SWE-bench ProNemotron 3 Ultra 550B A55B by 4.849.744.9+4.8
- SWE-bench VerifiedQwen3.5 122B A10B by 1.370.772+1.3
- SciCodetie40.339.7tie
AgentsQwen3.5 122B A10B
- BrowseCompQwen3.5 122B A10B by 19.444.463.8+19.4
- GDPValNemotron 3 Ultra 550B A55B by 1025.715.7+10
- τ-Bench V3 · BankingQwen3.5 122B A10B by 1.114.215.3+1.1
- Terminal-Bench 4.0tie0.50tie
ReasoningNemotron 3 Ultra 550B A55B
- Humanity's Last ExamNemotron 3 Ultra 550B A55B by 3.228.425.2+3.2
- CritPtNemotron 3 Ultra 550B A55B by 2.53.10.6+2.5
- GPQA DiamondNemotron 3 Ultra 550B A55B by 186.785.7+1
FactsEven
- AA-Omniscience · Non-hallucinationNemotron 3 Ultra 550B A55B by 57.470.312.9+57.4
- AA-Omniscience · AccuracyQwen3.5 122B A10B by 1.822.624.4+1.8
Long documentsNemotron 3 Ultra 550B A55B
- AA-LCRNemotron 3 Ultra 550B A55B by 379.376.3+3
- LongBench v2Nemotron 3 Ultra 550B A55B by 1.761.960.2+1.7
Following instructionsNemotron 3 Ultra 550B A55B
- IFBenchNemotron 3 Ultra 550B A55B by 5.781.475.7+5.7
- Multi-ChallengeNemotron 3 Ultra 550B A55B by 2.363.861.5+2.3
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.
- AA-OmniscienceNemotron 3 Ultra 550B A55B by 41.1-0.4-41.5+41.1
- GDPval-AA v2Nemotron 3 Ultra 550B A55B by 287 rating points1148861+287 rating
- tau2-bench retailNemotron 3 Ultra 550B A55B by 18.373.755.4+18.3
- tau2-bench airlineNemotron 3 Ultra 550B A55B by 16.379.863.5+16.3
- BrowseComp [offline]Qwen3.5 122B A10B by 144458+14
- LMArena v3 proxyQwen3.5 122B A10B by 12.734.347+12.7
- AA Agentic IndexNemotron 3 Ultra 550B A55B by 12.121.79.6+12.1
- τ²-Bench Telecom (AA run)Qwen3.5 122B A10B by 10.383.393.6+10.3
- AA IntelligenceNemotron 3 Ultra 550B A55B by 7.322.915.6+7.3
- WMT24++Nemotron 3 Ultra 550B A55B by 5.483.778.3+5.4
- HLE, text-onlyNemotron 3 Ultra 550B A55B by 529.224.2+5
- tau2-bench telecomQwen3.5 122B A10B by 4.28387.2+4.2
- τ³-Bench BankingNemotron 3 Ultra 550B A55B by 4.218.414.2+4.2
- Artificial Analysis Coding IndexNemotron 3 Ultra 550B A55B by 3.649.345.7+3.6
- DeepSWE [offline]Nemotron 3 Ultra 550B A55B by 2.630.4+2.6
- SciCode, subtaskNemotron 3 Ultra 550B A55B by 2.24239.8+2.2
- MMLU-ProXtie8382.2tie
- MMLU-Protie86.886.7tie
- Terminal Bench 4 [offline]tie00tie
Questions people ask
Which is better, Nemotron 3 Ultra 550B A55B or Qwen3.5 122B A10B?
Nemotron 3 Ultra 550B A55B wins four of the six areas where both have results: coding, reasoning, long documents and following instructions. Qwen3.5 122B A10B wins agents. They are level on facts.
Which is better for coding?
Nemotron 3 Ultra 550B A55B. It wins 5 of the 7 coding tests both models report; Qwen3.5 122B A10B wins 1, and 1 is a tie.
Which is cheaper?
Nemotron 3 Ultra 550B A55B costs $0.60 per million input tokens and $2.50 per million output tokens; Qwen3.5 122B A10B costs $0.40 and $3.20. That makes Nemotron 3 Ultra 550B A55B about 14% cheaper for the same work.
How do you compare the two?
We use the 39 benchmark tests both models have published scores on. The verdict counts the 20 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.