Capability
43.6/100
field median 57.3
#160 of 214 ranked models
Model profile · NVIDIA
Nemotron 3 Ultra
Nemotron 3 Ultra scores 43.6 out of 100 and ranks #160 of 214. This profile shows 19 source-displayable benchmark rows; its strongest eligible category is Instruction Following at #5. Published weights can be self-hosted, but infrastructure cost varies and is not a comparable API token rate.
Data as of July 31, 2026 · How the score is built
Strongest published evidence
Instruction Following ranks #5. A well-rounded choice across a range of tasks.
Validate before choosing
19 published rows leave some tracked benchmark slots empty. No comparable first-party API token rate is published.
Decision snapshot
Each value carries a field reference instead of floating alone. Markers compare this model with the current ranked and priced catalog; they are not absolute quality thresholds.
Price
Self-hosted; infrastructure cost varies
input median $1
No comparable first-party hosted token rate
Speed
Not measured
field median 108 tok/s
Time to first token not measured
Context
1Mtokens
field median 256,000
Reported for this model; direct source link not stored
Capability shape
Each axis shows percentile within that category’s eligible cohort. The comparison outline is the median of the six nearest public-score peers; a collapsed vertex means the category is not rank-eligible.
Nemotron 3 Ultra category percentile values
- Agentic8th percentile
- Coding25th percentile
- ReasoningNot eligible
- Knowledge9th percentile
- MathNot eligible
- Multilingual50th percentile
- MultimodalNot eligible
- Instruction following87th percentile
The dashed outline is median of 6 nearest peers.
Eligible category ranks
- Agentic#118/128
- Coding#97/129
- ReasoningNot ranked
- Knowledge#50/55
- MathNot ranked
- Multilingual#7/13
- MultimodalNot ranked
- Inst. Following#5/32
Top decileTop quartileMid-fieldNot eligible
How much of this is verified
Coverage is split by category so a strong number never hides a thin evidence base. Verified means the row is tied to a published source; provisional rows remain visible but separate.
- Agentic5/5 verified
- Coding5/5 verified
- Reasoning2/2 verified
- Knowledge5/5 verified
- MathNot measured
- Multilingual1/1 verified
- MultimodalNot measured
- Inst. Following1/1 verified
Verified sourceProvisionalNot measured
Spec sheet
Each documented value carries its source. Missing fields stay visible as not sourced or not published, rather than disappearing from the page.
- API model ID
- Not published
- Context window
- 1M
- Maximum output
- Not sourced yet
- Knowledge cutoff
- Not sourced yet
- Input modalities
- Not sourced yet
- Output modalities
- Not sourced yet
- Parameters
- Not sourced yet
- Availability
- Not sourced yet
- Cloud regions
- Not tracked yet
- Lifecycle
- Current
- API capabilities
- Tool calling, structured outputs, and batch support are not tracked yet
- Prompt caching
- Not documented in the pricing record
- Self-host
- Open weights available; hardware estimate not sourced
- Rate limits
- Not tracked yet
Deployment options
Self-host and provider-specific paths stay separate from benchmark evidence so operating constraints are visible before a score becomes the whole decision.
Published weights are available, but BenchLM does not yet have a sourced parameter and VRAM profile for this exact model. Hardware cost estimates stay unavailable until that sizing record is complete.
Estimate VRAM from known parametersCategory score record
Scores and ranks appear only where published evidence can be displayed. The table keeps the score, weight, cohort, and evidence state together.
| Category | Score | Rank | Percentile | Weight | Benchmarks | Evidence |
|---|---|---|---|---|---|---|
| AgenticRank #118 of 128Percentile 8thWeight 22%5 benchmarksVerified | 33.0 | #118 of 128 | 8th | 22% | 5 benchmarks | Verified |
| CodingRank #97 of 129Percentile 25thWeight 20%5 benchmarksVerified | 43.4 | #97 of 129 | 25th | 20% | 5 benchmarks | Verified |
| ReasoningRank Not rankedWeight 17%2 benchmarksVerified | 85.1 | Not ranked | Not available | 17% | 2 benchmarks | Verified |
| KnowledgeRank #50 of 55Percentile 9thWeight 12%5 benchmarksVerified | 53.1 | #50 of 55 | 9th | 12% | 5 benchmarks | Verified |
| MathWeight 5%0 benchmarksNot measured | Not measured | Not ranked | Not available | 5% | 0 benchmarks | Not measured |
| MultilingualRank #7 of 13Percentile 50thWeight 7%1 benchmarkVerified | 47.4 | #7 of 13 | 50th | 7% | 1 benchmark | Verified |
| MultimodalWeight 12%0 benchmarksNot measured | Not measured | Not ranked | Not available | 12% | 0 benchmarks | Not measured |
| Inst. FollowingRank #5 of 32Percentile 87thWeight 5%1 benchmarkVerified | 91.9 | #5 of 32 | 87th | 5% | 1 benchmark | Verified |
Benchmark ledger
Coding opens by default. The marker compares each value with the best source-verified result in the catalog; provisional leaders do not set the reference. Expand the remaining categories for every published row.
Coding5 rows
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| SWE-bench VerifiedSoftware Engineering Benchmark Verified | Score71.9% | Versus best verified row Best verified: Claude Opus 5 · 96% | Gap24.1 behind | WeightWeighted 16% | Provider exact |
| SciCodeScientific Code Benchmark | Score44.6% | Versus best verified row Best verified: Sakana Fugu · 60.1% | Gap15.5 behind | WeightWeighted 16% | Provider exact |
| SWE Multilingual | Score67.7% | Versus best verified row Best verified: Claude Opus 5 · 89.5% | Gap21.8 behind | WeightDisplay only | Provider exact |
| LiveCodeBench v6 | Score89.0% | Versus best verified row Best verified: Sakana Fugu-Ultra · 93.2% | Gap4.2 behind | WeightDisplay only | Provider exact |
| Terminal-Bench 2.0 | Score56.4% | Versus best verified row Best verified: GPT-5.6 Sol · 91.9% | Gap35.5 behind | WeightDisplay only | Provider exact |
Agentic5 rows
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| Terminal-Bench 2.0 | Score56.4% | Versus best verified row Best verified: GPT-5.6 Sol · 91.9% | Gap35.5 behind | WeightWeighted 38% | Provider exact |
| BrowseComp | Score44.4% | Versus best verified row Best verified: GPT-5.6 Sol · 92.2% | Gap47.8 behind | WeightWeighted 28% | Provider exact |
| PinchBench | Score90.0% | Versus best verified row Best verified: Nemotron 3 Ultra · 90.0% | GapBest verified | WeightDisplay only | Provider exact |
| τ³-bench resultsτ³-Bench Tool-Agent-User Evaluation | Score70.9% | Versus best verified row Best verified: Mistral Medium 3.5 128B · 91.4% | Gap20.5 behind | WeightDisplay only | Provider exact |
| HLE w/ toolsHumanity's Last Exam with tools | Score37.4% | Versus best verified row Best verified: Claude Opus 5 · 64.7% | Gap27.3 behind | WeightDisplay only | Provider exact |
Reasoning2 rows
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| LongBench v2 | Score61.9% | Versus best verified row Best verified: Qwen3.5 397B · 63.2% | Gap1.3 behind | WeightWeighted 38% | Provider exact |
| CritPtCritical Physics Tasks | Score3.1% | Versus best verified row Best verified: GLM-5.2 · 20.9% | Gap17.8 behind | WeightDisplay only | Provider exact |
Knowledge5 rows
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| HLEHumanity's Last Exam | Score26.7% | Versus best verified row Best verified: Claude Opus 5 · 64.7% | Gap38 behind | WeightWeighted 45% | Provider exact |
| MMLU-ProMassive Multitask Language Understanding Professional | Score86.8% | Versus best verified row Best verified: Qwen3.7 Max · 89.6% | Gap2.8 behind | WeightWeighted 30% | Provider exact |
| GPQAGraduate-Level Google-Proof Q&A | Score87% | Versus best verified row Best verified: Sakana Fugu-Ultra · 95.5% | Gap8.5 behind | WeightWeighted 7% | Provider exact |
| GPQA-DGPQA Diamond | Score87.0% | Versus best verified row Best verified: Sakana Fugu-Ultra · 95.5% | Gap8.5 behind | WeightDisplay only | Provider exact |
| HLE w/o toolsHumanity's Last Exam without tools | Score26.7% | Versus best verified row Best verified: Claude Mythos 5 · 59% | Gap32.3 behind | WeightDisplay only | Provider exact |
Multilingual1 row
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| MMLU-ProX | Score83% | Versus best verified row Best verified: Qwen3.7 Max · 87% | Gap4 behind | WeightWeighted 100% | Provider exact |
Inst. Following1 row
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| IFBenchInstruction Following Benchmark | Score81.7% | Versus best verified row Best verified: MAI-Thinking-1 · 85% | Gap3.3 behind | WeightWeighted 65% | Provider exact |
Lineage
The sequence follows explicit supersedes links. Scores and prices remain blank when the corresponding public row or first-party rate is unavailable.
Jun 4, 2026 · you are here
Nemotron 3 UltraScore 43.6 · Price not listed
Base entry
How to read this profile
The visual layer above carries the decisions. These notes preserve the model, ranking, coverage, and family context behind the numbers.
Nemotron 3 Ultra ranks #160 of 214 on the public leaderboard with a score of 43.56/100. It does not yet have enough sourced coverage for a verified position.
Nemotron 3 Ultra is a open weight model with a 1M context window. It uses an explicit reasoning mode, which can improve complex problem solving while adding latency and token use.
Official exact-value snapshot from the nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16 Hugging Face model card. BenchLM maps directly comparable values from NVIDIA's table and keeps provider-specific rows display-only where appropriate. NVIDIA's earlier announcement used approximate 500B/50B wording; the released checkpoint is 550B total / 55B active.
19 of 376 tracked benchmark slots currently have displayable evidence. Missing categories stay blank.
Its strongest eligible category is Instruction Following at #5, while its lowest eligible position is Agentic at #118. a well-rounded choice across a range of tasks.
Radar
Full release historyNemotron 3 Ultra release history
Nemotron 3 Ultra released
NVIDIA · Model release
Source confirmed
Frequently asked questions
How does Nemotron 3 Ultra perform overall in AI benchmarks?
Nemotron 3 Ultra ranks #160 out of 214 models on the public BenchAlign leaderboard, with a score of 43.56/100. Its evidence status is Estimated, and this profile shows 19 source-displayable benchmark rows. The label describes evidence depth, not a provider quality claim; inspect category rows before choosing a workload.
Is Nemotron 3 Ultra good for knowledge and understanding?
Nemotron 3 Ultra ranks #50 out of 55 eligible models for knowledge and understanding, with a public category score of 53.1/100. Higher-ranked alternatives are available for workloads where this category decides the choice. Check the underlying rows before treating the aggregate as a workload guarantee.
Is Nemotron 3 Ultra good for coding and programming?
Nemotron 3 Ultra ranks #97 out of 129 eligible models for coding and programming, with a public category score of 43.4/100. Higher-ranked alternatives are available for workloads where this category decides the choice. Check the underlying rows before treating the aggregate as a workload guarantee.
Is Nemotron 3 Ultra good for reasoning and logic?
Nemotron 3 Ultra has source-displayable benchmark coverage for reasoning and logic, but the public category table does not assign it a rank there. The individual rows remain available for inspection. A missing category position means the evidence threshold was not met; it does not convert the model's unmeasured work into a zero.
Is Nemotron 3 Ultra good for agentic tool use and computer tasks?
Nemotron 3 Ultra ranks #118 out of 128 eligible models for agentic tool use and computer tasks, with a public category score of 33/100. Higher-ranked alternatives are available for workloads where this category decides the choice. Check the underlying rows before treating the aggregate as a workload guarantee.
Is Nemotron 3 Ultra good for instruction following?
Nemotron 3 Ultra ranks #5 out of 32 eligible models for instruction following, with a public category score of 91.9/100. That places it in the current top ten for this category. Check the underlying rows before treating the aggregate as a workload guarantee.
Is Nemotron 3 Ultra good for multilingual tasks?
Nemotron 3 Ultra ranks #7 out of 13 eligible models for multilingual tasks, with a public category score of 47.4/100. That places it in the current top ten for this category. Check the underlying rows before treating the aggregate as a workload guarantee.
Is Nemotron 3 Ultra open source?
Nemotron 3 Ultra is an open-weight model from NVIDIA. Its weights can be downloaded for local or hosted deployment, subject to the published license. Open weight does not automatically mean open source: training data and training code may remain private, and commercial restrictions can still apply.
Does Nemotron 3 Ultra have full benchmark coverage on BenchLM?
No. Nemotron 3 Ultra currently has 39 source-displayable rows across 376 tracked benchmark slots. The profile exposes published, non-generated evidence and leaves missing categories blank until an exact evaluation is available. Coverage describes how much was measured; it is not a penalty added to an individual benchmark result.
What is the context window size of Nemotron 3 Ultra?
Nemotron 3 Ultra has a reported context window of 1M in the exact-model catalog record. The value stays visible, but the profile marks its source link as unavailable instead of presenting it as directly documented. Maximum output length remains separate because providers often publish a different limit.
Last updated July 31, 2026. Runtime fields remain blank until a sourced snapshot exists.
Watch Nemotron 3 Ultra in the weekly brief
Get one weekly email when material rank, price, availability, or benchmark evidence changes are worth revisiting.
Read a sample issueJoin 2,000+ readers.