Agentic
Like-for-like- Claude Opus 5
- 77.4
- Supported · #2/151
- Nemotron 3 Ultra
- 22.2
- Supported · #150/151
- Basis
- BenchAlign lane · 19 vs 6 public rows
- Reading
- Claude Opus 5 leads
Claude Opus 5 vs Nemotron 3 Ultra
Updated September 4, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.
Decision reading
Claude Opus 5 has the higher public score, 80.66 versus 40.67, and the 90% score intervals do not overlap.
11 results are shared. Category rows resting on Estimated evidence or different benchmark sets are marked directional and do not name a winner.
Share or export
Which one for your work
Recommendations appear only when a shared evidence basis or an explicit operating constraint supports the call. Secondary and unsupported use cases stay disclosed below the initial list.
Coding work
Code generation, repair, and software-engineering tasks
Claude Opus 5
Claude Opus 5 leads on the public coding lane, 75.6 to 26.9, with Supported evidence for both models and non-overlapping 90% intervals.
Confidence: stronger
Agentic work
Tool use, computer use, and multi-step task completion
Claude Opus 5
Claude Opus 5 leads on the public agentic lane, 77.4 to 22.2, with Supported evidence for both models and non-overlapping 90% intervals.
Confidence: stronger
Show secondary and unsupported calls
Long documents
Prompts that approach the documented context limit
No clear pick
The documented context windows are equal.
Confidence: documented
Chat turn cost
1K fresh input + 500 output tokens
Not enough matched evidence
A complete comparable API-rate estimate is not available for both models.
Confidence: listed-rates
Cache-heavy agent loop cost
200K cached + 20K fresh input + 10K output tokens
Not enough matched evidence
A complete comparable API-rate estimate is not available for both models.
Confidence: listed-rates
Repository review cost
50K fresh input + 3K output tokens
Not enough matched evidence
A complete comparable API-rate estimate is not available for both models.
Confidence: listed-rates
What is actually comparable
Shared results can support a head-to-head reading. Results present for only one model describe coverage, not superiority.
62 Claude Opus 5 only
11 Shared
13 Nemotron 3 Ultra only
- Shared results
- 11
- Claude Opus 5 only
- 62
- Nemotron 3 Ultra only
- 13
- Like-for-like categories
- 2 / 8
1 category rests on Estimated evidence or different evidence sets. Those rows remain visible for coverage context but do not name a winner.
Category results, on a stated basis
Each row shows the public-lane category score for both models: the BenchAlign lane for agentic, coding, and knowledge, and the weighted public rows elsewhere, with the same rank each model page reports. A row is like-for-like only when both scores rest on Supported evidence or the same weighted set. Directional and not-comparable rows remain visible, but they never receive a winner.
Coding
Like-for-like- Claude Opus 5
- 75.6
- Supported · #3/183
- Nemotron 3 Ultra
- 26.9
- Supported · #174/183
- Basis
- BenchAlign lane · 16 vs 7 public rows
- Reading
- Claude Opus 5 leads
Knowledge
Directional only- Claude Opus 5
- 82.1
- Supported · #3/181
- Nemotron 3 Ultra
- 45.6
- Estimated · #114/181
- Basis
- BenchAlign lane · 19 vs 7 public rows
- Reading
- Directional only
Reasoning
Not comparable- Claude Opus 5
- 75.6
- #13/22
- Nemotron 3 Ultra
- 48.7
- Unranked · 1 rankable row
- Basis
- Provisional lane · 2 vs 1 weighted rows
- Reading
- Not comparable
Math
Not comparable- Claude Opus 5
- Not ranked
- Nemotron 3 Ultra
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Multilingual
Not comparable- Claude Opus 5
- Not ranked
- Nemotron 3 Ultra
- 47.4
- #7/12
- Basis
- Provisional lane · 0 vs 1 weighted rows
- Reading
- Not comparable
Multimodal
Not comparable- Claude Opus 5
- 88.7
- #2/48
- Nemotron 3 Ultra
- Not ranked
- Basis
- Provisional lane · 1 vs 0 weighted rows
- Reading
- Not comparable
Instruction following
Not comparable- Claude Opus 5
- Not ranked
- Nemotron 3 Ultra
- 88.7
- #28/120
- Basis
- Provisional lane · 0 vs 1 weighted rows
- Reading
- Not comparable
| Category | Claude Opus 5 | Nemotron 3 Ultra | Basis | Reading |
|---|---|---|---|---|
| Agentic | 77.4Supported · #2/151 | 22.2Supported · #150/151 | Like-for-likeBenchAlign lane · 19 vs 6 public rows | Claude Opus 5 leads |
| Coding | 75.6Supported · #3/183 | 26.9Supported · #174/183 | Like-for-likeBenchAlign lane · 16 vs 7 public rows | Claude Opus 5 leads |
| Knowledge | 82.1Supported · #3/181 | 45.6Estimated · #114/181 | Directional onlyBenchAlign lane · 19 vs 7 public rows | Directional only |
| Reasoning | 75.6#13/22 | 48.7Unranked · 1 rankable row | Not comparableProvisional lane · 2 vs 1 weighted rows | Not comparable |
| Math | Not ranked | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Multilingual | Not ranked | 47.4#7/12 | Not comparableProvisional lane · 0 vs 1 weighted rows | Not comparable |
| Multimodal | 88.7#2/48 | Not ranked | Not comparableProvisional lane · 1 vs 0 weighted rows | Not comparable |
| Instruction following | Not ranked | 88.7#28/120 | Not comparableProvisional lane · 0 vs 1 weighted rows | Not comparable |
Ranks count the models scored in each category’s lane, so the agentic, coding, and knowledge denominators (BenchAlign) differ from the provisional-lane categories. Unranked scores sit on the lane’s scale but never name a winner. Like-for-like rows within 0.5 points read as a practical tie.
Shape of the matched evidence
Only shared public evidence is shown. Sparse evidence stays a ruled list rather than being closed into a radar shape.
Too few matched category axes support a radar. The ruled list below shows only shared benchmark results; positions use each benchmark’s normalized display scale when available.
BrowseComp
Agentic
Claude Opus 5: 90.8%Nemotron 3 Ultra: 44.4%Normalized gap 46.4Claude Opus 5 source Nemotron 3 Ultra sourceHLE
Knowledge
Claude Opus 5: 64.7%Nemotron 3 Ultra: 26.7%Normalized gap 38.0Claude Opus 5 source Nemotron 3 Ultra sourceHLE w/o tools
Knowledge
Claude Opus 5: 56.3%Nemotron 3 Ultra: 26.7%Normalized gap 29.6Claude Opus 5 source Nemotron 3 Ultra sourceSWE-bench Verified
Coding
Claude Opus 5: 96%Nemotron 3 Ultra: 71.9%Normalized gap 24.1Claude Opus 5 source Nemotron 3 Ultra sourceSWE Multilingual
Coding
Claude Opus 5: 89.5%Nemotron 3 Ultra: 67.7%Normalized gap 21.8Claude Opus 5 source Nemotron 3 Ultra source
What each workload costs
Three fixed token mixes turn per-token rates into comparable decisions. Each scenario states context fit and whether cached input had to fall back to the published list-input rate.
Chat turn
1K fresh input + 500 output tokens
- Claude Opus 5
- $0.0175
- Fits in one request
- Nemotron 3 Ultra
- Self-hosted; infrastructure cost varies
- Fits in one request
Nemotron 3 Ultra has no comparable published API token rate.
Repository review
50K fresh input + 3K output tokens
- Claude Opus 5
- $0.325
- Fits in one request
- Nemotron 3 Ultra
- Self-hosted; infrastructure cost varies
- Fits in one request
Nemotron 3 Ultra has no comparable published API token rate.
Cache-heavy agent loop
200K cached + 20K fresh input + 10K output tokens
- Claude Opus 5
- $0.45
- Fits in one request
- Nemotron 3 Ultra
- Self-hosted; infrastructure cost varies
- Fits in one request
- Cached-input rate unavailable
Nemotron 3 Ultra has no comparable published API token rate.
Specification differences
Sourced differences are shown directly. Missing facts stay explicit instead of being inferred from a model name or family.
SpecificationClaude Opus 5Nemotron 3 Ultra
Context window
Maximum documented context; output-token limits may be lower.
Claude Opus 5
Nemotron 3 Ultra
1M
API model ID
Claude Opus 5
claude-opus-5
Anthropic model overviewNemotron 3 Ultra
Not sourced
Cached-input rate
A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.
Claude Opus 5
$0.5 per 1M cached input tokens
Claude API pricingNemotron 3 Ultra
No comparable hosted API rate
Documented inputs
Claude Opus 5
text, image
Anthropic model overviewNemotron 3 Ultra
Not sourced
Documented outputs
Claude Opus 5
Nemotron 3 Ultra
Not sourced
Provider availability
Claude Opus 5
Generally Available · Claude API
Anthropic model overviewNemotron 3 Ultra
Not sourced
Reasoning profile
Claude Opus 5
Reasoning
Nemotron 3 Ultra
Reasoning
Weight access
Claude Opus 5
Proprietary
Nemotron 3 Ultra
Open Weight
License
Claude Opus 5
Proprietary
Nemotron 3 Ultra
Open Weight
Release date
Claude Opus 5
2026-07-24
Nemotron 3 Ultra
2026-06-04
If you already use one of these models
- Deployment change
- The models list different providers, so authentication, endpoint behavior, limits, and feature support may change.
- Quality signal
- Claude Opus 5 has the higher public score, 80.66 versus 40.67, and the 90% score intervals do not overlap.
- Workload cost
- A complete comparable API-rate estimate is not available for both models.
- Context tradeoff
- Both models list 1M.
Run the same representative tasks against both endpoints before changing production traffic.
Benchmark evidence
The full public result ledger is available for audit without forcing a wide desktop table onto a phone.
Browse raw public benchmark evidence86 rows
Agentic
Terminal-Bench 3.0
Claude Opus 542.7%SourceNemotron 3 Ultra—Not directly comparable
BrowseComp
Claude Opus 590.8%SourceNemotron 3 Ultra44.4%SourceClaude Opus 5 leads this result
HLE w/ tools
Claude Opus 564.7%SourceNemotron 3 Ultra37.4%SourceClaude Opus 5 leads this result
DeepSearchQA
Claude Opus 595.0%SourceNemotron 3 Ultra—Not directly comparable
DRACO
Claude Opus 588.6%SourceNemotron 3 Ultra—Not directly comparable
BrowseComp (10-agent, prerelease)
Claude Opus 593.6%SourceNemotron 3 Ultra—Not directly comparable
OSWorld 2.0
Claude Opus 570.6%SourceNemotron 3 Ultra—Not directly comparable
MCP Atlas
Claude Opus 585.8%SourceNemotron 3 Ultra—Not directly comparable
MCP-Atlas claim coverage
Claude Opus 589.1%SourceNemotron 3 Ultra—Not directly comparable
LAB all-pass (Anthropic harness)
Claude Opus 523.58%SourceNemotron 3 Ultra—Not directly comparable
LAB criterion-pass (Anthropic harness)
Claude Opus 593.74%SourceNemotron 3 Ultra—Not directly comparable
LAB all-pass (Harvey held-out)
Claude Opus 511.7%SourceNemotron 3 Ultra—Not directly comparable
LAB criterion-pass (Harvey held-out)
Claude Opus 594.1%SourceNemotron 3 Ultra—Not directly comparable
Toolathlon-Verified
Claude Opus 580.6%SourceNemotron 3 Ultra—Not directly comparable
Toolathlon Verified Pass@3
Claude Opus 587.0%SourceNemotron 3 Ultra—Not directly comparable
Toolathlon Verified Pass³
Claude Opus 573.1%SourceNemotron 3 Ultra—Not directly comparable
Toolathlon Verified avg. turns
Claude Opus 523.5 turnsSourceNemotron 3 Ultra—Not directly comparable
AutomationBench
Claude Opus 526.0%SourceNemotron 3 Ultra—Not directly comparable
Terminal-Bench 2.1 (Vals)
Claude Opus 584.6%SourceNemotron 3 Ultra50.9%SourceClaude Opus 5 leads this result
Terminal-Bench 2.0
Claude Opus 5—Nemotron 3 Ultra56.4%SourceNot directly comparable
PinchBench
Claude Opus 5—Nemotron 3 Ultra90.0%SourceNot directly comparable
τ³-bench results
Claude Opus 5—Nemotron 3 Ultra70.9%SourceNot directly comparable
Coding
Bug Hunt Bench
Claude Opus 527 fixesSourceNemotron 3 Ultra—Not directly comparable
SWE-bench Verified
Claude Opus 596%SourceNemotron 3 Ultra71.9%SourceClaude Opus 5 leads this result
SWE-bench Pro
Claude Opus 579.2%SourceNemotron 3 Ultra—Not directly comparable
SWE Multilingual
Claude Opus 589.5%SourceNemotron 3 Ultra67.7%SourceClaude Opus 5 leads this result
SWE Multimodal
Claude Opus 559.4%SourceNemotron 3 Ultra—Not directly comparable
deepSwe
Claude Opus 568.8%SourceNemotron 3 Ultra—Not directly comparable
FrontierCode 1.1 Main
Claude Opus 553.4%SourceNemotron 3 Ultra—Not directly comparable
FrontierCode 1.1 Extended
Claude Opus 563.6%SourceNemotron 3 Ultra—Not directly comparable
FrontierSWE v2
Claude Opus 552.0%SourceNemotron 3 Ultra—Not directly comparable
ProgramBench (episode 1)
Claude Opus 583.0%SourceNemotron 3 Ultra—Not directly comparable
ProgramBench
Claude Opus 593.0%SourceNemotron 3 Ultra—Not directly comparable
cursorBench32
Claude Opus 570.0%SourceNemotron 3 Ultra—Not directly comparable
VulcanBench v3
Claude Opus 587.0%SourceNemotron 3 Ultra—Not directly comparable
VulcanBench CII v1
Claude Opus 596.4%SourceNemotron 3 Ultra—Not directly comparable
LiveCodeBench (Vals)
Claude Opus 589.0%SourceNemotron 3 Ultra86.0%SourceClaude Opus 5 leads this result
SWE-bench (Vals)
Claude Opus 597.0%SourceNemotron 3 Ultra69.0%SourceClaude Opus 5 leads this result
LiveCodeBench v6
Claude Opus 5—Nemotron 3 Ultra89.0%SourceNot directly comparable
SciCode
Claude Opus 5—Nemotron 3 Ultra44.6%SourceNot directly comparable
Terminal-Bench 2.0
Claude Opus 5—Nemotron 3 Ultra56.4%SourceNot directly comparable
Reasoning
ARC-AGI-1
Claude Opus 597.50%SourceNemotron 3 Ultra—Not directly comparable
ARC-AGI-2
Claude Opus 590.4%SourceNemotron 3 Ultra—Not directly comparable
ARC-AGI-3
Claude Opus 530.2%SourceNemotron 3 Ultra—Not directly comparable
CritPt
Claude Opus 5—Nemotron 3 Ultra3.1%SourceNot directly comparable
LongBench v2
Claude Opus 5—Nemotron 3 Ultra61.9%SourceNot directly comparable
Knowledge
HLE
Claude Opus 564.7%SourceNemotron 3 Ultra26.7%SourceClaude Opus 5 leads this result
HLE w/o tools
Claude Opus 556.3%SourceNemotron 3 Ultra26.7%SourceClaude Opus 5 leads this result
HLE-Verified
Claude Opus 554.4%SourceNemotron 3 Ultra—Not directly comparable
LABBench2
Claude Opus 584.2%SourceNemotron 3 Ultra—Not directly comparable
HealthBench (raw)
Claude Opus 567.1%SourceNemotron 3 Ultra—Not directly comparable
HealthBench (length-adjusted)
Claude Opus 557.8%SourceNemotron 3 Ultra—Not directly comparable
HealthBench Professional
Claude Opus 559.8%SourceNemotron 3 Ultra—Not directly comparable
HealthBench Professional (raw)
Claude Opus 573.4%SourceNemotron 3 Ultra—Not directly comparable
BioMysteryBench (human-solvable)
Claude Opus 590.1%SourceNemotron 3 Ultra—Not directly comparable
BioMysteryBench (human-difficult)
Claude Opus 549.4%SourceNemotron 3 Ultra—Not directly comparable
SpatialBench Verified
Claude Opus 572.5%SourceNemotron 3 Ultra—Not directly comparable
SingleCellBench
Claude Opus 560.6%SourceNemotron 3 Ultra—Not directly comparable
ProteinGym Hard
Claude Opus 547.7%SourceNemotron 3 Ultra—Not directly comparable
Protein Design
Claude Opus 542.5%SourceNemotron 3 Ultra—Not directly comparable
Organic chemistry V2
Claude Opus 561.6%SourceNemotron 3 Ultra—Not directly comparable
Protocols (troubleshooting)
Claude Opus 561.1%SourceNemotron 3 Ultra—Not directly comparable
Protocols (understanding)
Claude Opus 578.4%SourceNemotron 3 Ultra—Not directly comparable
GPQA Diamond (Vals)
Claude Opus 593.4%SourceNemotron 3 Ultra86.1%SourceClaude Opus 5 leads this result
MMLU-Pro (Vals)
Claude Opus 591.6%SourceNemotron 3 Ultra85.8%SourceClaude Opus 5 leads this result
GPQA
Claude Opus 5—Nemotron 3 Ultra87%SourceNot directly comparable
GPQA-D
Claude Opus 5—Nemotron 3 Ultra87.0%SourceNot directly comparable
MMLU-Pro
Claude Opus 5—Nemotron 3 Ultra86.8%SourceNot directly comparable
Math
IMO 2026
Claude Opus 542/42SourceNemotron 3 Ultra—Not directly comparable
RiemannBench (no tools)
Claude Opus 560.0%SourceNemotron 3 Ultra—Not directly comparable
RiemannBench (tools)
Claude Opus 579.0%SourceNemotron 3 Ultra—Not directly comparable
ArXivMath Jun. 2026 (no tools)
Claude Opus 590.8%SourceNemotron 3 Ultra—Not directly comparable
ArXivMath Jun. 2026 (tools)
Claude Opus 591.3%SourceNemotron 3 Ultra—Not directly comparable
Multilingual
Multimodal
Chartography (no tools)
Claude Opus 529.6%SourceNemotron 3 Ultra—Not directly comparable
Chartography (tools)
Claude Opus 583.0%SourceNemotron 3 Ultra—Not directly comparable
BenchCAD Vision2Code (no tools)
Claude Opus 50.366SourceNemotron 3 Ultra—Not directly comparable
BenchCAD Vision2Code (tools)
Claude Opus 50.821SourceNemotron 3 Ultra—Not directly comparable
GDP.pdf (no tools)
Claude Opus 583.4%SourceNemotron 3 Ultra—Not directly comparable
GDP.pdf (tools)
Claude Opus 585.5%SourceNemotron 3 Ultra—Not directly comparable
OfficeQA
Claude Opus 578.1%SourceNemotron 3 Ultra—Not directly comparable
OfficeQA Pro
Claude Opus 566.9%SourceNemotron 3 Ultra—Not directly comparable
Instruction following
IFBench
Claude Opus 5—Nemotron 3 Ultra81.7%SourceNot directly comparable
Frequently asked questions
Which is better, Claude Opus 5 or Nemotron 3 Ultra?
Claude Opus 5 has the higher public score, 80.66 versus 40.67, and the 90% score intervals do not overlap. That is the clearest overall quality signal in the current public data.
Which is better for coding, Claude Opus 5 or Nemotron 3 Ultra?
Claude Opus 5 leads the public coding lane, 75.6 to 26.9, with Supported evidence for both models and non-overlapping 90% intervals.
Which is better for agentic tasks, Claude Opus 5 or Nemotron 3 Ultra?
Claude Opus 5 leads the public agentic tasks lane, 77.4 to 22.2, with Supported evidence for both models and non-overlapping 90% intervals.
Which costs less, Claude Opus 5 or Nemotron 3 Ultra?
Both models do not have comparable published API token rates, so this page does not name a universal price winner.
Which has the larger context window, Claude Opus 5 or Nemotron 3 Ultra?
Both models list the same context window, 1M.
Related comparisons
Last updated September 4, 2026
Watch Claude Opus 5 vs Nemotron 3 Ultra
One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.
Read a sample issueJoin 2,000+ readers.