Enter opens the highlighted result. Esc closes.

GLM 5.2 vs Nemotron 3 Ultra

GLM 5.2 leads Nemotron 3 Ultra on coding, 61.1 to 30.0 on the Vibe Coding Index, and Nemotron 3 Ultra costs 1.8x less per output token ($2.40 vs $4.40 per million). Updated October 23, 2026; 71 benchmarks have a result for both.

Replace GLM 5.2 Replace Nemotron 3 Ultra
GLM 5.2Z.ai
Coding
61.1
Overall
62.5
Output $/M
$4.40
64
Nemotron 3 UltraNVIDIA
Coding
30.0
Overall
53.1
Output $/M
$2.40
7
vs boards won of 71 shared

GLM 5.2 or Nemotron 3 Ultra: which to pick

Pick GLM 5.2 for coding: it scores 61.1 on the Vibe Coding Index against 30.0 for Nemotron 3 Ultra.

Pick Nemotron 3 Ultra for output tokens 1.8x cheaper ($2.40 vs $4.40 per million) and work measured by LiveCodeBench (Vals AI) or GPQA Diamond, where it leads.

Model AGLM 5.2Z.ai61.1/1004 independent benchmarks$0.86 per solved taskModel BNemotron 3 UltraNVIDIA30.0/1003 independent benchmarksNo cost per solved task
Shared results
71
GLM 5.2 only
33
Nemotron 3 Ultra only
0
Like-for-like categories
8 of 9

Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.

Which one for your work

Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.

At a glance

Coding index
GLM 5.2 61.1 from 4 benchmarks, Nemotron 3 Ultra 30.0 from 3 benchmarks.
Widest category
Agentic work, 18.4 points apart on average over 7 shared benchmarks.
Widest single gap
Vibe Code Bench (Vals AI): GLM 5.2 64.0% against Nemotron 3 Ultra 7.6%, 56.3 points apart on a 0 to 100 scale.
Output price
GLM 5.2 $4.40 against Nemotron 3 Ultra $2.40 per million output tokens, 1.8x apart.
Context window
GLM 5.2 takes 1.05M tokens, Nemotron 3 Ultra 262K.

Benchmark by benchmark

First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.

Boards won, where both have a result
CategoryGLM 5.2BarsNemotron 3 UltraTies
Coding121
Intelligence20
Agentic work70
Reasoning62
Math31
Knowledge90
Web design80
General173
All shared boards647
CodingLike-for-likeSix of these boards feed the Vibe Coding Index121
Coding benchmarks for GLM 5.2 and Nemotron 3 Ultra
BenchmarkGLM 5.2Nemotron 3 UltraLeader
LMArena WebDev1604Not measuredOnly GLM 5.2 measured
AA Coding Index68.849.3GLM 5.2
LiveCodeBench (Vals AI)69.5%86.0%Nemotron 3 Ultra
SWE-bench Verified (Vals AI)82.8%69.0%GLM 5.2
Vibe Code Bench (Vals AI)64.0%7.6%GLM 5.2
Code migration (Vals AI)37.9%4.9%GLM 5.2
ProgramBench (Vals AI)0.5%0.0%GLM 5.2
Terminal-Bench 2.1 (Vals AI)67.8%50.9%GLM 5.2
SkillsBench (Vals AI)45.1%Not measuredOnly GLM 5.2 measured
SciCode51.2%40.3%GLM 5.2
Terminal-Bench Hard (Artificial Analysis run)50.8%36.4%GLM 5.2
FrontierCode24.5%13.6%GLM 5.2
WeirdML70.1%43.5%GLM 5.2
ALE-Bench1047.0Not measuredOnly GLM 5.2 measured
LMArena Text: coding14871467GLM 5.2
LMArena Text: software and IT14861468GLM 5.2
LMArena WebDev: React1617Not measuredOnly GLM 5.2 measured
LMArena WebDev: HTML1567Not measuredOnly GLM 5.2 measured
SWE Atlas: Codebase QnA48.1%Not measuredOnly GLM 5.2 measured
SWE Atlas: Refactoring42.4%Not measuredOnly GLM 5.2 measured
SWE Atlas: Test Writing41.5%Not measuredOnly GLM 5.2 measured
Neither model has a result on the other 22 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
Intelligence benchmarks for GLM 5.2 and Nemotron 3 Ultra
BenchmarkGLM 5.2Nemotron 3 UltraLeader
AA Intelligence Index33.722.9GLM 5.2
Epoch Capabilities Index151.9146.0GLM 5.2
Agentic workLike-for-likePublic boards, shown as published, kept out of the index70
Agentic work benchmarks for GLM 5.2 and Nemotron 3 Ultra
BenchmarkGLM 5.2Nemotron 3 UltraLeader
AA Agentic Index38.420.1GLM 5.2
Design Arena: agents building web apps1222Not measuredOnly GLM 5.2 measured
Design Arena: agents building full-stack apps1241Not measuredOnly GLM 5.2 measured
Design Arena: agents building mobile apps1160Not measuredOnly GLM 5.2 measured
Design Arena: agents building Android apps1179Not measuredOnly GLM 5.2 measured
Design Arena: agents building Godot games1142Not measuredOnly GLM 5.2 measured
Design Arena: agents building games1206Not measuredOnly GLM 5.2 measured
Design Arena: HTML slides1080Not measuredOnly GLM 5.2 measured
Design Arena: PowerPoint slides1157Not measuredOnly GLM 5.2 measured
SREBench (Vals AI)0.0%Not measuredOnly GLM 5.2 measured
Finance Agent (Vals AI)49.7%37.7%GLM 5.2
Legal Research Bench (Vals AI)31.3%15.4%GLM 5.2
Excel Modeling Benchmark (Vals AI)61.5%31.7%GLM 5.2
τ²-Bench Telecom99.1%83.3%GLM 5.2
APEX-Agents45.2%22.7%GLM 5.2
PostTrainBench31.7%Not measuredOnly GLM 5.2 measured
Vending-Bench 2$8,314Not measuredOnly GLM 5.2 measured
MCP Atlas77.8%63.1%GLM 5.2
Neither model has a result on the other 24 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index62
Reasoning benchmarks for GLM 5.2 and Nemotron 3 Ultra
BenchmarkGLM 5.2Nemotron 3 UltraLeader
GPQA Diamond85.6%86.1%Nemotron 3 Ultra
GPQA Diamond (Artificial Analysis run)89.5%86.7%GLM 5.2
Humanity's Last Exam (Artificial Analysis run)41.1%28.4%GLM 5.2
GPQA Diamond (Epoch AI run)91.9%85.4%GLM 5.2
ARC-AGI77.0%Not measuredOnly GLM 5.2 measured
ARC-AGI-222.8%Not measuredOnly GLM 5.2 measured
SimpleBench58.8%Not measuredOnly GLM 5.2 measured
Chess puzzles22.0%Not measuredOnly GLM 5.2 measured
Mystery game puzzles19.0%20.0%Nemotron 3 Ultra
EBR-bench9.5%Not measuredOnly GLM 5.2 measured
LMArena Text: hard prompts14801452GLM 5.2
LMArena Text: expert14871472GLM 5.2
LMArena Text: hard prompts in English14811453GLM 5.2
Kagi LLM Benchmark62.6%Not measuredOnly GLM 5.2 measured
Neither model has a result on the other 9 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index31
Math benchmarks for GLM 5.2 and Nemotron 3 Ultra
BenchmarkGLM 5.2Nemotron 3 UltraLeader
OTIS Mock AIME86.4%86.7%Nemotron 3 Ultra
FrontierMath, tiers 1 to 359.2%Not measuredOnly GLM 5.2 measured
FrontierMath, tier 429.3%Not measuredOnly GLM 5.2 measured
ProofBench35.0%2.0%GLM 5.2
LMArena Text: math14841457GLM 5.2
LMArena Text: mathematical work14901462GLM 5.2
MathArena22.7%Not measuredOnly GLM 5.2 measured
Neither model has a result on the other 7 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index90
Knowledge benchmarks for GLM 5.2 and Nemotron 3 Ultra
BenchmarkGLM 5.2Nemotron 3 UltraLeader
MMLU Pro86.7%85.8%GLM 5.2
LegalBench (Vals AI)84.1%82.1%GLM 5.2
CorpFin (Vals AI)66.1%65.5%GLM 5.2
TaxEval (Vals AI)73.3%73.1%GLM 5.2
MedCode (Vals AI)40.8%38.6%GLM 5.2
SimpleQA Verified34.2%Not measuredOnly GLM 5.2 measured
LMArena Text: science14891465GLM 5.2
LMArena Text: medicine and healthcare14751455GLM 5.2
LMArena Text: legal and government14861453GLM 5.2
LMArena Text: business and finance14511435GLM 5.2
Neither model has a result on the other 10 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index80
Web design benchmarks for GLM 5.2 and Nemotron 3 Ultra
BenchmarkGLM 5.2Nemotron 3 UltraLeader
Design Arena: websites12851145GLM 5.2
Design Arena: UI components12941135GLM 5.2
Design Arena: game dev12671138GLM 5.2
Design Arena: data viz12861142GLM 5.2
Design Arena: 3D12881132GLM 5.2
Design Arena: all code categories12871146GLM 5.2
Design Arena: SVG12131069GLM 5.2
Design Arena: ASCII art11911100GLM 5.2
Neither model has a result on the other 1 board here
GeneralLike-for-likePublic boards, shown as published, kept out of the index173
General benchmarks for GLM 5.2 and Nemotron 3 Ultra
BenchmarkGLM 5.2Nemotron 3 UltraLeader
MedScribe (Vals AI)83.5%Not measuredOnly GLM 5.2 measured
IFBench73.3%81.4%Nemotron 3 Ultra
Long context reasoning (AA-LCR)78.3%79.3%Nemotron 3 Ultra
DTBench93.6%90.1%GLM 5.2
LMCA45.8%36.9%GLM 5.2
Surface Evolver Bench55.6%Not measuredOnly GLM 5.2 measured
LMArena Text14701445GLM 5.2
LMArena Text: instruction following14661418GLM 5.2
LMArena Text: creative writing14621398GLM 5.2
LMArena Text: English14751454GLM 5.2
LMArena Text: non-English14581427GLM 5.2
LMArena Text: Chinese15221498GLM 5.2
LMArena Text: German14711474Nemotron 3 Ultra
LMArena Text: Japanese1449Not measuredOnly GLM 5.2 measured
LMArena Text: Korean14381385GLM 5.2
LMArena Text: French14781473GLM 5.2
LMArena Text: Spanish14781454GLM 5.2
LMArena Text: Russian14671417GLM 5.2
LMArena Text: Polish14571385GLM 5.2
LMArena Text: multi-turn14691411GLM 5.2
LMArena Text: longer queries14791437GLM 5.2
LMArena Text: writing and language14631405GLM 5.2
LMArena Text: entertainment and media14491397GLM 5.2
EQ-Bench 41222Not measuredOnly GLM 5.2 measured
Neither model has a result on the other 24 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for GLM 5.2 and Nemotron 3 Ultra
BenchmarkGLM 5.2Nemotron 3 UltraLeader
Neither model has a result on 5 boards here

What each workload costs

Three fixed token mixes at each model's listed price, so per-token rates become a bill.

Chat turn

1K input + 500 output tokens

GLM 5.2$0.0036Fits in one request
Nemotron 3 Ultra$0.0018Fits in one request

Nemotron 3 Ultra costs less

Repository review

50K input + 3K output tokens

GLM 5.2$0.083Fits in one request
Nemotron 3 Ultra$0.037Fits in one request

Nemotron 3 Ultra costs less

Agent loop

200K input + 10K output tokens

GLM 5.2$0.324Fits in one request
Nemotron 3 Ultra$0.144Fits in one request

Nemotron 3 Ultra costs less

Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.

Specification differences

The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.

Specifications of GLM 5.2 and Nemotron 3 Ultra
SpecificationZ.ai GLM 5.2NVIDIA Nemotron 3 Ultra
MakerZ.aiNVIDIA
WeightsOpen weightsOpen weights
ReleasedJune 16, 2026June 4, 2026
Context window1.05M262K
Input $/Mtok$1.40$0.60
Output $/Mtok$4.40$2.40
Coding index61.130.0
Independent measurements behind it43
Cost per solved task$0.86Not computable
SourceOpenRouter listingOpenRouter listing

Questions

Is GLM 5.2 better than Nemotron 3 Ultra for coding?

Pick GLM 5.2 for coding: it scores 61.1 on the Vibe Coding Index against 30.0 for Nemotron 3 Ultra. Pick Nemotron 3 Ultra for output tokens 1.8x cheaper ($2.40 vs $4.40 per million) and work measured by LiveCodeBench (Vals AI) or GPQA Diamond, where it leads.

Which is cheaper, GLM 5.2 or Nemotron 3 Ultra?

GLM 5.2 costs $1.40 per million input tokens and $4.40 per million output tokens; Nemotron 3 Ultra costs $0.60 and $2.40. Cost per solved task is known for GLM 5.2 only.

Which has the bigger context window, GLM 5.2 or Nemotron 3 Ultra?

GLM 5.2, with 1.05M tokens against 262K.

How many benchmarks is the GLM 5.2 vs Nemotron 3 Ultra comparison based on?

71 benchmarks have a result for both models as of October 23, 2026. GLM 5.2's coding index rests on 4 independent benchmarks and Nemotron 3 Ultra's on 3. Where only one model has a result, the page says so instead of naming a winner.

Compare next

Embed this comparison on your site

Free under CC BY 4.0, while the source link stays visible. Preview the frame · Terms and other embeds

<iframe src="https://llmbenchmark.ai/embed/compare/glm-5-2-vs-nvidia-nemotron-3-ultra-550b-a55b" title="GLM 5.2 vs Nemotron 3 Ultra, LLMBenchmark.ai" width="100%" height="360" style="border:0;max-width:100%" loading="lazy"></iframe>
<p style="font:13px/1.4 sans-serif;margin:6px 0 0">Source: <a href="https://llmbenchmark.ai/compare/glm-5-2-vs-nvidia-nemotron-3-ultra-550b-a55b">LLMBenchmark.ai, GLM 5.2 vs Nemotron 3 Ultra</a></p>