GPT-5.4 vs Nemotron 3 Ultra 550B A55B
Wins 6 of 7 areas
Coding · Agents · Reasoning · Facts · Math · Long documents
Wins 1 of 7 areas
Following instructions
GPT-5.4 is the stronger all-rounder.Nemotron 3 Ultra 550B A55B is cheaper and better at following instructions.
Scores updated · 36 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 software70GPT-5.47 of 7 tests
- AgentsCarrying out multi-step tasks on its own40GPT-5.44 of 4 tests
- ReasoningHard problems that need careful thinking40GPT-5.44 of 4 tests
- MathCompetition and research-level math30GPT-5.43 of 4 tests · 1 tie
- FactsGetting facts right instead of making them up21GPT-5.42 of 3 tests
- Long documentsFinding answers in very long texts10GPT-5.41 of 1 test
- Following instructionsDoing exactly what it is asked02Nemotron 3 Ultra 550B A55B2 of 2 tests
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.
Nemotron 3 Ultra 550B A55B costs 82% less for the same work.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where GPT-5.4 pulls ahead
- Finds hard-to-locate facts by browsing the webBrowseComp+38.3points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+28.3points ahead
- Customer-service tasks in a simulated bankτ-Bench V3 · Banking+25.4points ahead
Where Nemotron 3 Ultra 550B A55B pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+62points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+7.5points ahead
- Follows detailed instructions exactlyLiveBench · Instruction Following+3.3points ahead
Every test, side by side
All 36 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGPT-5.4
- Terminal-Bench 2.1GPT-5.4 by 24.478.353.9+24.4
- Terminal-Bench HardGPT-5.4 by 21.257.636.4+21.2
- SciCodeGPT-5.4 by 16.356.640.3+16.3
- LiveBench · Agentic CodingGPT-5.4 by 15.153.838.7+15.1
- SWE-bench ProGPT-5.4 by 857.749.7+8
- LiveBench · CodingGPT-5.4 by 6.877.570.7+6.8
- SWE-bench MultilingualGPT-5.4 by 471.767.7+4
AgentsGPT-5.4
- BrowseCompGPT-5.4 by 38.382.744.4+38.3
- τ-Bench V3 · BankingGPT-5.4 by 25.439.614.2+25.4
- MCP AtlasGPT-5.4 by 22.567.244.7+22.5
- GDPValGPT-5.4 by 11.637.325.7+11.6
ReasoningGPT-5.4
- CritPtGPT-5.4 by 20.323.43.1+20.3
- Humanity's Last ExamGPT-5.4 by 15.343.728.4+15.3
- LiveBench · ReasoningGPT-5.4 by 13.488.174.7+13.4
- GPQA DiamondGPT-5.4 by 5.39286.7+5.3
FactsGPT-5.4
- AA-Omniscience · Non-hallucinationNemotron 3 Ultra 550B A55B by 628.370.3+62
- AA-Omniscience · AccuracyGPT-5.4 by 28.350.922.6+28.3
- SimpleQA VerifiedGPT-5.4 by 12.745.132.4+12.7
MathGPT-5.4
- HMMT Feb. 2026GPT-5.4 by 18.997.778.8+18.9
- LiveBench · MathematicsGPT-5.4 by 5.594.288.7+5.5
- AIME 2026GPT-5.4 by 599.294.2+5
- IMOAnswerBenchtie91.492.3tie
Following instructionsNemotron 3 Ultra 550B A55B
- IFBenchNemotron 3 Ultra 550B A55B by 7.573.981.4+7.5
- LiveBench · Instruction FollowingNemotron 3 Ultra 550B A55B by 3.370.273.5+3.3
Other results11 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.
- Apex (Pass@1)Nemotron 3 Ultra 550B A55B by 30.754.184.8+30.7
- livebench_data_analysisGPT-5.4 by 24.879.354.5+24.8
- AA Agentic IndexGPT-5.4 by 22.544.221.7+22.5
- Artificial Analysis Coding IndexGPT-5.4 by 21.871.149.3+21.8
- AA IntelligenceGPT-5.4 by 16.13922.9+16.1
- HLE (with tools)GPT-5.4 by 14.752.137.4+14.7
- livebench_languageGPT-5.4 by 11.882.670.8+11.8
- AA-OmniscienceGPT-5.4 by 6.25.8-0.4+6.2
- τ²-Bench Telecom (AA run)GPT-5.4 by 3.887.183.3+3.8
- Finance AgentGPT-5.4 by 2.35653.7+2.3
- MMLU-Protie87.586.8tie
Questions people ask
Which is better, GPT-5.4 or Nemotron 3 Ultra 550B A55B?
GPT-5.4 wins six of the seven areas where both have results: coding, agents, reasoning, facts, math and long documents. Nemotron 3 Ultra 550B A55B wins following instructions, and costs 82% less.
Which is better for coding?
GPT-5.4. It wins 7 of the 7 coding tests both models report; Nemotron 3 Ultra 550B A55B wins none.
Which is cheaper?
GPT-5.4 costs $2.50 per million input tokens and $15.00 per million output tokens; Nemotron 3 Ultra 550B A55B costs $0.60 and $2.50. That makes Nemotron 3 Ultra 550B A55B about 82% cheaper for the same work.
How do you compare the two?
We use the 36 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 11 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.