GLM-5.2 vs Nemotron 3 Ultra 550B A55B
Wins 5 of 7 areas
Coding · Agents · Reasoning · Facts · Math
Wins 2 of 7 areas
Long documents · Following instructions
GLM-5.2 is the stronger all-rounder.Nemotron 3 Ultra 550B A55B is cheaper and better at following instructions.
Scores updated · 39 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 software60GLM-5.26 of 6 tests
- ReasoningHard problems that need careful thinking50GLM-5.25 of 5 tests
- AgentsCarrying out multi-step tasks on its own30GLM-5.23 of 4 tests · 1 tie
- FactsGetting facts right instead of making them up30GLM-5.23 of 3 tests
- MathCompetition and research-level math31GLM-5.23 of 4 tests
- Following instructionsDoing exactly what it is asked02Nemotron 3 Ultra 550B A55B2 of 2 tests
- Long documentsFinding answers in very long texts01Nemotron 3 Ultra 550B A55B1 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.
Nemotron 3 Ultra 550B A55B costs 43% less for the same work.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where GLM-5.2 pulls ahead
- Multi-step tasks using many real software toolsMCP Atlas+32.1points ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+24points ahead
- Customer-service tasks in a simulated bankτ-Bench V3 · Banking+20.4points ahead
Where Nemotron 3 Ultra 550B A55B pulls ahead
- Follows detailed instructions exactlyLiveBench · Instruction Following+11.2points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+8.1points ahead
- Olympiad math problems with short answersIMOAnswerBench+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.
CodingGLM-5.2
- Terminal-Bench 2.1GLM-5.2 by 2477.953.9+24
- Terminal-Bench HardGLM-5.2 by 14.450.836.4+14.4
- LiveBench · Agentic CodingGLM-5.2 by 13.151.838.7+13.1
- SWE-bench ProGLM-5.2 by 12.462.149.7+12.4
- SciCodeGLM-5.2 by 10.951.240.3+10.9
- LiveBench · CodingGLM-5.2 by 979.770.7+9
AgentsGLM-5.2
- MCP AtlasGLM-5.2 by 32.176.844.7+32.1
- τ-Bench V3 · BankingGLM-5.2 by 20.434.614.2+20.4
- GDPValGLM-5.2 by 17.943.625.7+17.9
- Terminal-Bench 4.0tie10.5tie
ReasoningGLM-5.2
- CritPtGLM-5.2 by 17.820.93.1+17.8
- SimpleBenchGLM-5.2 by 17.158.841.7+17.1
- Humanity's Last ExamGLM-5.2 by 12.741.128.4+12.7
- LiveBench · ReasoningGLM-5.2 by 3.978.674.7+3.9
- GPQA DiamondGLM-5.2 by 2.889.586.7+2.8
FactsGLM-5.2
- AA-Omniscience · Non-hallucinationGLM-5.2 by 3.473.770.3+3.4
- SimpleQA VerifiedGLM-5.2 by 1.834.232.4+1.8
- AA-Omniscience · AccuracyGLM-5.2 by 1.724.322.6+1.7
MathGLM-5.2
- HMMT Feb. 2026GLM-5.2 by 13.792.578.8+13.7
- AIME 2026GLM-5.2 by 599.294.2+5
- IMOAnswerBenchNemotron 3 Ultra 550B A55B by 1.39192.3+1.3
- LiveBench · MathematicsGLM-5.2 by 1.189.888.7+1.1
Long documentsNemotron 3 Ultra 550B A55B
- AA-LCRNemotron 3 Ultra 550B A55B by 178.379.3+1
Following instructionsNemotron 3 Ultra 550B A55B
- LiveBench · Instruction FollowingNemotron 3 Ultra 550B A55B by 11.262.373.5+11.2
- IFBenchNemotron 3 Ultra 550B A55B by 8.173.381.4+8.1
Other results14 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.
- GDPval-AA v2GLM-5.2 by 360 rating points15081148+360 rating
- Toolathlon VerifiedGLM-5.2 by 25.659.934.3+25.6
- Artificial Analysis Coding IndexGLM-5.2 by 19.568.849.3+19.5
- livebench_data_analysisGLM-5.2 by 19.273.754.5+19.2
- AA Agentic IndexGLM-5.2 by 17.739.421.7+17.7
- HLE (with tools)GLM-5.2 by 17.354.737.4+17.3
- τ²-Bench Telecom (AA run)GLM-5.2 by 15.899.183.3+15.8
- SWEBench Pro PublicGLM-5.2 by 15.762.146.4+15.7
- AA IntelligenceGLM-5.2 by 10.833.722.9+10.8
- τ³-Bench BankingGLM-5.2 by 8.426.818.4+8.4
- livebench_languageGLM-5.2 by 5.476.270.8+5.4
- AA-OmniscienceGLM-5.2 by 4.84.4-0.4+4.8
- Global-MMLU-LiteGLM-5.2 by 3.689.285.6+3.6
- StrongREJECTtie98.598.7tie
Questions people ask
Which is better, GLM-5.2 or Nemotron 3 Ultra 550B A55B?
GLM-5.2 wins five of the seven areas where both have results: coding, agents, reasoning, facts and math. Nemotron 3 Ultra 550B A55B wins long documents and following instructions, and costs 43% less.
Which is better for coding?
GLM-5.2. It wins 6 of the 6 coding tests both models report; Nemotron 3 Ultra 550B A55B wins none.
Which is cheaper?
GLM-5.2 costs $1.40 per million input tokens and $4.40 per million output tokens; Nemotron 3 Ultra 550B A55B costs $0.60 and $2.70. That makes Nemotron 3 Ultra 550B A55B about 43% 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 25 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 14 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.