Enter opens the highlighted result. Esc closes.

GLM 4.7 vs Nemotron 3 Ultra

GLM 4.7 leads Nemotron 3 Ultra on coding, 55.1 to 30.0 on the Vibe Coding Index, at about the same output price ($2.20 vs $2.40 per million). Updated October 22, 2026; 56 benchmarks have a result for both.

Replace GLM 4.7 Replace Nemotron 3 Ultra
GLM 4.7Z.ai
Coding
55.1
Overall
56.9
Output $/M
$2.20
20
Nemotron 3 UltraNVIDIA
Coding
30.0
Overall
53.1
Output $/M
$2.40
36
vs boards won of 56 shared

GLM 4.7 or Nemotron 3 Ultra: which to pick

Pick GLM 4.7 for coding: it scores 55.1 on the Vibe Coding Index against 30.0 for Nemotron 3 Ultra, though on 2 benchmarks against 3.

Pick Nemotron 3 Ultra for a larger context window (262K against 205K tokens) and work measured by AA Coding Index or LiveCodeBench (Vals AI), where it leads.

Model AGLM 4.7Z.ai55.1/1002 independent benchmarks$0.29 per solved taskModel BNemotron 3 UltraNVIDIA30.0/1003 independent benchmarksNo cost per solved task
Shared results
56
GLM 4.7 only
26
Nemotron 3 Ultra only
15
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 4.7 55.1 from 2 benchmarks, Nemotron 3 Ultra 30.0 from 3 benchmarks.
Widest category
Agentic work, 8.8 points apart on average over 2 shared benchmarks.
Widest single gap
CorpFin (Vals AI): GLM 4.7 46.4% against Nemotron 3 Ultra 65.5%, 19.1 points apart on a 0 to 100 scale.
Output price
GLM 4.7 $2.20 against Nemotron 3 Ultra $2.40 per million output tokens, 1.1x apart.
Context window
GLM 4.7 takes 205K 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 4.7BarsNemotron 3 UltraTies
Coding15
Intelligence02
Agentic work11
Reasoning07
Math13
Knowledge27
Web design80
General711
All shared boards2036
CodingLike-for-likeSix of these boards feed the Vibe Coding Index15
Coding benchmarks for GLM 4.7 and Nemotron 3 Ultra
BenchmarkGLM 4.7Nemotron 3 UltraLeader
LMArena WebDev1435Not measuredOnly GLM 4.7 measured
AA Coding Index45.349.3Nemotron 3 Ultra
LiveCodeBench (Vals AI)82.2%86.0%Nemotron 3 Ultra
SWE-bench Verified (Vals AI)69.4%69.0%GLM 4.7
Vibe Code Bench (Vals AI)Not measured7.6%Only Nemotron 3 Ultra measured
Code migration (Vals AI)Not measured4.9%Only Nemotron 3 Ultra measured
ProgramBench (Vals AI)Not measured0.0%Only Nemotron 3 Ultra measured
Terminal-Bench 2.1 (Vals AI)Not measured50.9%Only Nemotron 3 Ultra measured
LiveCodeBench (Artificial Analysis run)89.4%Not measuredOnly GLM 4.7 measured
SciCodeNot measured40.3%Only Nemotron 3 Ultra measured
Terminal-Bench Hard (Artificial Analysis run)31.8%36.4%Nemotron 3 Ultra
Terminal-Bench33.4%Not measuredOnly GLM 4.7 measured
FrontierCodeNot measured13.6%Only Nemotron 3 Ultra measured
WeirdMLNot measured43.5%Only Nemotron 3 Ultra measured
ALE-Bench399.5Not measuredOnly GLM 4.7 measured
LMArena Text: coding14541468Nemotron 3 Ultra
LMArena Text: software and IT14531468Nemotron 3 Ultra
LMArena WebDev: React1420Not measuredOnly GLM 4.7 measured
LMArena WebDev: HTML1447Not measuredOnly GLM 4.7 measured
Neither model has a result on the other 24 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index02
Intelligence benchmarks for GLM 4.7 and Nemotron 3 Ultra
BenchmarkGLM 4.7Nemotron 3 UltraLeader
AA Intelligence Index22.222.9Nemotron 3 Ultra
Epoch Capabilities Index143.6146.0Nemotron 3 Ultra
Agentic workLike-for-likePublic boards, shown as published, kept out of the index11
Agentic work benchmarks for GLM 4.7 and Nemotron 3 Ultra
BenchmarkGLM 4.7Nemotron 3 UltraLeader
AA Agentic IndexNot measured20.1Only Nemotron 3 Ultra measured
Design Arena: agents building full-stack apps1050Not measuredOnly GLM 4.7 measured
Design Arena: agents building mobile apps1105Not measuredOnly GLM 4.7 measured
Design Arena: agents building Android apps1112Not measuredOnly GLM 4.7 measured
Design Arena: agents building Godot games1058Not measuredOnly GLM 4.7 measured
Finance Agent (Vals AI)Not measured37.7%Only Nemotron 3 Ultra measured
Legal Research Bench (Vals AI)Not measured15.4%Only Nemotron 3 Ultra measured
Excel Modeling Benchmark (Vals AI)Not measured31.7%Only Nemotron 3 Ultra measured
τ²-Bench Telecom95.9%83.3%GLM 4.7
APEX-AgentsNot measured22.7%Only Nemotron 3 Ultra measured
Vending-Bench 2$2,377Not measuredOnly GLM 4.7 measured
BrowseComp-Plus64.7%Not measuredOnly GLM 4.7 measured
MCP Atlas58.1%63.1%Nemotron 3 Ultra
Neither model has a result on the other 29 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index07
Reasoning benchmarks for GLM 4.7 and Nemotron 3 Ultra
BenchmarkGLM 4.7Nemotron 3 UltraLeader
GPQA Diamond80.0%86.1%Nemotron 3 Ultra
GPQA Diamond (Artificial Analysis run)85.9%86.7%Nemotron 3 Ultra
Humanity's Last Exam (Artificial Analysis run)27.4%28.4%Nemotron 3 Ultra
GPQA Diamond (Epoch AI run)83.3%85.4%Nemotron 3 Ultra
SimpleBench47.7%Not measuredOnly GLM 4.7 measured
Chess puzzles21.0%Not measuredOnly GLM 4.7 measured
Mystery game puzzlesNot measured20.0%Only Nemotron 3 Ultra measured
LMArena Text: hard prompts14431452Nemotron 3 Ultra
LMArena Text: expert14241472Nemotron 3 Ultra
LMArena Text: hard prompts in English14521453Nemotron 3 Ultra
Neither model has a result on the other 13 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index13
Math benchmarks for GLM 4.7 and Nemotron 3 Ultra
BenchmarkGLM 4.7Nemotron 3 UltraLeader
AIME93.3%Not measuredOnly GLM 4.7 measured
MGSM (Vals AI)88.2%Not measuredOnly GLM 4.7 measured
AA Math Index95.0Not measuredOnly GLM 4.7 measured
AIME 2025 (Artificial Analysis run)95.0%Not measuredOnly GLM 4.7 measured
OTIS Mock AIME83.3%86.7%Nemotron 3 Ultra
ProofBench6.0%2.0%GLM 4.7
LMArena Text: math14241457Nemotron 3 Ultra
LMArena Text: mathematical work14111461Nemotron 3 Ultra
Neither model has a result on the other 6 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index27
Knowledge benchmarks for GLM 4.7 and Nemotron 3 Ultra
BenchmarkGLM 4.7Nemotron 3 UltraLeader
MMLU Pro82.7%85.8%Nemotron 3 Ultra
LegalBench (Vals AI)83.4%82.1%GLM 4.7
CaseLaw (Vals AI)54.9%Not measuredOnly GLM 4.7 measured
CorpFin (Vals AI)46.4%65.5%Nemotron 3 Ultra
TaxEval (Vals AI)68.8%73.1%Nemotron 3 Ultra
MedQA (Vals AI)93.7%Not measuredOnly GLM 4.7 measured
MedCode (Vals AI)32.8%38.6%Nemotron 3 Ultra
MMLU Pro (Artificial Analysis run)85.6%Not measuredOnly GLM 4.7 measured
SimpleQA Verified32.2%Not measuredOnly GLM 4.7 measured
LMArena Text: science14571465Nemotron 3 Ultra
LMArena Text: medicine and healthcare14691456GLM 4.7
LMArena Text: legal and government14441452Nemotron 3 Ultra
LMArena Text: business and finance14291434Nemotron 3 Ultra
Neither model has a result on the other 7 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index80
Web design benchmarks for GLM 4.7 and Nemotron 3 Ultra
BenchmarkGLM 4.7Nemotron 3 UltraLeader
Design Arena: websites12251145GLM 4.7
Design Arena: UI components11971135GLM 4.7
Design Arena: game dev11881138GLM 4.7
Design Arena: data viz11961141GLM 4.7
Design Arena: 3D11901131GLM 4.7
Design Arena: all code categories12161146GLM 4.7
Design Arena: SVG11421069GLM 4.7
Design Arena: ASCII art11651100GLM 4.7
Neither model has a result on the other 1 board here
GeneralLike-for-likePublic boards, shown as published, kept out of the index711
General benchmarks for GLM 4.7 and Nemotron 3 Ultra
BenchmarkGLM 4.7Nemotron 3 UltraLeader
MedScribe (Vals AI)68.6%Not measuredOnly GLM 4.7 measured
IFBench67.9%81.4%Nemotron 3 Ultra
Long context reasoning (AA-LCR)71.0%79.3%Nemotron 3 Ultra
DTBenchNot measured90.1%Only Nemotron 3 Ultra measured
LMCANot measured36.9%Only Nemotron 3 Ultra measured
CL-bench15.9%Not measuredOnly GLM 4.7 measured
CL-bench Life10.9%Not measuredOnly GLM 4.7 measured
LMArena Text14351445Nemotron 3 Ultra
LMArena Text: instruction following14121418Nemotron 3 Ultra
LMArena Text: creative writing14011399GLM 4.7
LMArena Text: English14531455Nemotron 3 Ultra
LMArena Text: non-English14171427Nemotron 3 Ultra
LMArena Text: Chinese14951497Nemotron 3 Ultra
LMArena Text: German14261472Nemotron 3 Ultra
LMArena Text: Japanese1439Not measuredOnly GLM 4.7 measured
LMArena Text: Korean13991384GLM 4.7
LMArena Text: French14321472Nemotron 3 Ultra
LMArena Text: Spanish14341455Nemotron 3 Ultra
LMArena Text: Russian14231417GLM 4.7
LMArena Text: Polish14351385GLM 4.7
LMArena Text: multi-turn14461411GLM 4.7
LMArena Text: longer queries14321437Nemotron 3 Ultra
LMArena Text: writing and language14111406GLM 4.7
LMArena Text: entertainment and media14001397GLM 4.7
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 4.7 and Nemotron 3 Ultra
BenchmarkGLM 4.7Nemotron 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 4.7$0.0017Fits in one request
Nemotron 3 Ultra$0.0018Fits in one request

GLM 4.7 costs less

Repository review

50K input + 3K output tokens

GLM 4.7$0.037Fits in one request
Nemotron 3 Ultra$0.037Fits in one request

GLM 4.7 costs less

Agent loop

200K input + 10K output tokens

GLM 4.7$0.142Exceeds the context window
Nemotron 3 Ultra$0.144Fits in one request

GLM 4.7 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 4.7 and Nemotron 3 Ultra
SpecificationZ.ai GLM 4.7NVIDIA Nemotron 3 Ultra
MakerZ.aiNVIDIA
WeightsOpen weightsOpen weights
ReleasedDecember 22, 2025June 4, 2026
Context window205K262K
Input $/Mtok$0.60$0.60
Output $/Mtok$2.20$2.40
Coding index55.130.0
Independent measurements behind it23
Cost per solved task$0.29Not computable
SourceOpenRouter listingOpenRouter listing

Questions

Is GLM 4.7 better than Nemotron 3 Ultra for coding?

Pick GLM 4.7 for coding: it scores 55.1 on the Vibe Coding Index against 30.0 for Nemotron 3 Ultra, though on 2 benchmarks against 3. Pick Nemotron 3 Ultra for a larger context window (262K against 205K tokens) and work measured by AA Coding Index or LiveCodeBench (Vals AI), where it leads.

Which is cheaper, GLM 4.7 or Nemotron 3 Ultra?

GLM 4.7 costs $0.60 per million input tokens and $2.20 per million output tokens; Nemotron 3 Ultra costs $0.60 and $2.40. Cost per solved task is known for GLM 4.7 only.

Which has the bigger context window, GLM 4.7 or Nemotron 3 Ultra?

Nemotron 3 Ultra, with 262K tokens against 205K.

How many benchmarks is the GLM 4.7 vs Nemotron 3 Ultra comparison based on?

56 benchmarks have a result for both models as of October 22, 2026. GLM 4.7's coding index rests on 2 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-4-7-vs-nvidia-nemotron-3-ultra-550b-a55b" title="GLM 4.7 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-4-7-vs-nvidia-nemotron-3-ultra-550b-a55b">LLMBenchmark.ai, GLM 4.7 vs Nemotron 3 Ultra</a></p>