Coding
Like-for-like- Mercury 2
- 24.6
- Supported · #149/151
- Nemotron 3 Ultra
- 26.8
- Supported · #145/151
- Basis
- BenchAlign lane · 0 vs 7 public rows
- Reading
- Nemotron 3 Ultra leads · intervals overlap
Mercury 2 vs Nemotron 3 Ultra
Updated September 10, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.
Decision reading
The public evidence has no benchmark result shared by both models, so it does not support a quality verdict. Use the documented cost, context, and runtime rows instead.
0 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
Nemotron 3 Ultra
Nemotron 3 Ultra leads on the public coding lane, 26.8 to 24.6, with Supported evidence for both models, although the 90% intervals overlap.
Confidence: limited
Long documents
Prompts that approach the documented context limit
Nemotron 3 Ultra
Nemotron 3 Ultra has the larger documented context window.
Confidence: documented
Show secondary and unsupported calls
Agentic work
Tool use, computer use, and multi-step task completion
Not enough matched evidence
Mercury 2 is scored on Estimated evidence for agentic, so the reading is directional rather than like-for-like.
Confidence: limited
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
The page does not recommend a cost winner because at least one model cannot fit the stated workload in one request. Mercury 2 does not fit this workload in one request. Nemotron 3 Ultra has no comparable published API token rate.
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.
24 Nemotron 3 Ultra only
- Shared results
- 0
- Mercury 2 only
- 0
- Nemotron 3 Ultra only
- 24
- Like-for-like categories
- 1 / 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.
Agentic
Directional only- Mercury 2
- 38.1
- Estimated · #124/152
- Nemotron 3 Ultra
- 27.1
- Supported · #142/152
- Basis
- BenchAlign lane · 0 vs 6 public rows
- Reading
- Directional only
Reasoning
Not comparable- Mercury 2
- Not ranked
- Nemotron 3 Ultra
- 48.7
- Unranked · 1 rankable row
- Basis
- Provisional lane · 0 vs 1 weighted rows
- Reading
- Not comparable
Knowledge
Not comparable- Mercury 2
- Not ranked
- Nemotron 3 Ultra
- 45.0
- Estimated · #112/182
- Basis
- BenchAlign lane · 0 vs 7 public rows
- Reading
- Not comparable
Math
Not comparable- Mercury 2
- Not ranked
- Nemotron 3 Ultra
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Multilingual
Not comparable- Mercury 2
- Not ranked
- Nemotron 3 Ultra
- 47.4
- #7/12
- Basis
- Provisional lane · 0 vs 1 weighted rows
- Reading
- Not comparable
Multimodal
Not comparable- Mercury 2
- Not ranked
- Nemotron 3 Ultra
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Instruction following
Not comparable- Mercury 2
- Not ranked
- Nemotron 3 Ultra
- 88.8
- #28/121
- Basis
- Provisional lane · 0 vs 1 weighted rows
- Reading
- Not comparable
| Category | Mercury 2 | Nemotron 3 Ultra | Basis | Reading |
|---|---|---|---|---|
| Coding | 24.6Supported · #149/151 | 26.8Supported · #145/151 | Like-for-likeBenchAlign lane · 0 vs 7 public rows | Nemotron 3 Ultra leads · intervals overlap |
| Agentic | 38.1Estimated · #124/152 | 27.1Supported · #142/152 | Directional onlyBenchAlign lane · 0 vs 6 public rows | Directional only |
| Reasoning | Not ranked | 48.7Unranked · 1 rankable row | Not comparableProvisional lane · 0 vs 1 weighted rows | Not comparable |
| Knowledge | Not ranked | 45.0Estimated · #112/182 | Not comparableBenchAlign lane · 0 vs 7 public 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 | Not ranked | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Instruction following | Not ranked | 88.8#28/121 | 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.
A shared-evidence shape is not available.
BenchLM does not draw a radar or infer missing axes when the matched evidence is too sparse.
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
- Mercury 2
- $0.00063
- 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
- Mercury 2
- $0.01475
- 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
- Mercury 2
- $0.0175
- Does not fit in one request
- Nemotron 3 Ultra
- Self-hosted; infrastructure cost varies
- Fits in one request
- Cached-input rate unavailable
Mercury 2 does not fit this workload in one request. 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.
SpecificationMercury 2Nemotron 3 Ultra
Context window
Maximum documented context; output-token limits may be lower.
Mercury 2
128K
Nemotron 3 Ultra
1M
API model ID
Mercury 2
Not sourced
Nemotron 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.
Mercury 2
$0.025 per 1M cached input tokens
Nemotron 3 Ultra
No comparable hosted API rate
Documented inputs
Mercury 2
Not sourced
Nemotron 3 Ultra
Not sourced
Documented outputs
Mercury 2
Not sourced
Nemotron 3 Ultra
Not sourced
Provider availability
Mercury 2
Not sourced
Nemotron 3 Ultra
Not sourced
Reasoning profile
Mercury 2
Reasoning
Nemotron 3 Ultra
Reasoning
Weight access
Mercury 2
Proprietary
Nemotron 3 Ultra
Open Weight
License
Mercury 2
Proprietary
Nemotron 3 Ultra
Open Weight
Release date
Mercury 2
2026-02-24
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
- The public evidence has no benchmark result shared by both models, so it does not support a quality verdict.
- Workload cost
- A complete comparable API-rate estimate is not available for both models.
- Context tradeoff
- Nemotron 3 Ultra has the larger documented window (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 evidence24 rows
Agentic
Terminal-Bench 2.0
Mercury 2—Nemotron 3 Ultra56.4%SourceNot directly comparable
PinchBench
Mercury 2—Nemotron 3 Ultra90.0%SourceNot directly comparable
BrowseComp
Mercury 2—Nemotron 3 Ultra44.4%SourceNot directly comparable
τ³-bench results
Mercury 2—Nemotron 3 Ultra70.9%SourceNot directly comparable
HLE w/ tools
Mercury 2—Nemotron 3 Ultra37.4%SourceNot directly comparable
Terminal-Bench 2.1 (Vals)
Mercury 2—Nemotron 3 Ultra50.9%SourceNot directly comparable
Coding
SWE-bench Verified
Mercury 2—Nemotron 3 Ultra71.9%SourceNot directly comparable
SWE Multilingual
Mercury 2—Nemotron 3 Ultra67.7%SourceNot directly comparable
LiveCodeBench v6
Mercury 2—Nemotron 3 Ultra89.0%SourceNot directly comparable
SciCode
Mercury 2—Nemotron 3 Ultra44.6%SourceNot directly comparable
Terminal-Bench 2.0
Mercury 2—Nemotron 3 Ultra56.4%SourceNot directly comparable
LiveCodeBench (Vals)
Mercury 2—Nemotron 3 Ultra86.0%SourceNot directly comparable
SWE-bench (Vals)
Mercury 2—Nemotron 3 Ultra69.0%SourceNot directly comparable
Reasoning
Knowledge
GPQA
Mercury 2—Nemotron 3 Ultra87%SourceNot directly comparable
GPQA-D
Mercury 2—Nemotron 3 Ultra87.0%SourceNot directly comparable
HLE
Mercury 2—Nemotron 3 Ultra26.7%SourceNot directly comparable
HLE w/o tools
Mercury 2—Nemotron 3 Ultra26.7%SourceNot directly comparable
MMLU-Pro
Mercury 2—Nemotron 3 Ultra86.8%SourceNot directly comparable
GPQA Diamond (Vals)
Mercury 2—Nemotron 3 Ultra86.1%SourceNot directly comparable
MMLU-Pro (Vals)
Mercury 2—Nemotron 3 Ultra85.8%SourceNot directly comparable
Multilingual
MMLU-ProX
Mercury 2—Nemotron 3 Ultra83%SourceNot directly comparable
Instruction following
IFBench
Mercury 2—Nemotron 3 Ultra81.7%SourceNot directly comparable
Frequently asked questions
Which is better, Mercury 2 or Nemotron 3 Ultra?
The public evidence has no benchmark result shared by both models, so it does not support a quality verdict. The page therefore keeps the decision tied to the specific documented workload.
Which is better for coding, Mercury 2 or Nemotron 3 Ultra?
Nemotron 3 Ultra leads the public coding lane, 26.8 to 24.6, with Supported evidence for both models, although the 90% intervals overlap.
Which is better for agentic tasks, Mercury 2 or Nemotron 3 Ultra?
Mercury 2 scores higher for agentic tasks on the public lane, 38.1 to 27.1. Mercury 2 is scored on Estimated evidence for agentic tasks, so the reading is directional rather than like-for-like. BenchLM does not name a winner for agentic tasks from a directional reading; read the shared benchmark rows directly and test the models on the same task set.
Which costs less, Mercury 2 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, Mercury 2 or Nemotron 3 Ultra?
Nemotron 3 Ultra has the larger documented context window: 1M, compared with 128K.
Related comparisons
Last updated September 10, 2026
Watch Mercury 2 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.