Model comparison
Nemotron Ultra 253B vs Nova Pro
Head-to-head evidence from 11 shared benchmark results across 5 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.
Public leaderboard positions: Nemotron Ultra 253B #150 (Estimated); Nova Pro #196 (Supported). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.
Evidence parity. Nemotron Ultra 253B and Nova Pro share 11 comparable benchmark results. 0 of 8 categories are comparable. 0 results are unique to Nemotron Ultra 253B; 1 to Nova Pro.
Updated July 22, 2026- Shared results
- 11
- Nemotron Ultra 253B only
- 0
- Nova Pro only
- 1
- Comparable categories
- 0 / 8
Benchmark data for Nemotron Ultra 253B and Nova Pro is coming soon on BenchLM.
Confidence note. This is a partial-evidence comparison with 11 shared benchmark results across 5 evidence categories; 0 of 8 categories currently have scoreable aggregates for both models. Treat the verdict as directional until coverage is more balanced.
Why this result
BenchLM has partial data for these models, but not enough overlapping benchmark coverage to produce a fair score-level comparison yet.
Nova Pro has the larger context window at 128K, compared with 32K for Nemotron Ultra 253B.
Category breakdown
Exact category averages are shown below. Not measured means BenchLM does not have enough sourced public coverage for that model and category.
Operational comparison
Runtime and commercial metrics are compared only when both models have a complete sourced value.
| Metric | Nemotron Ultra 253B | Nova Pro | Comparison |
|---|---|---|---|
| Input / output priceUSD per 1M tokens | Nemotron Ultra 253B$0 input / $0 output | Nova ProNot available | A complete price comparison is not available. |
| Generation speedtokens per second | Nemotron Ultra 253BNot available | Nova Pro141 tok/s | A complete speed comparison is not available. |
| First-answer latencyseconds to first token | Nemotron Ultra 253BNot available | Nova Pro0.81 s | A complete latency comparison is not available. |
| Context windowmaximum listed tokens | Nemotron Ultra 253B32K | Nova Pro128K | Nova Pro lists the larger context window. |
Benchmark Deep Dive
Agentic1 benchmarks
| Benchmark | Nemotron Ultra 253B | Nova Pro | Result |
|---|---|---|---|
| τ²-bench resultsSource | 11.4% | 14% | Nova Pro leads |
Coding1 benchmarks
| Benchmark | Nemotron Ultra 253B | Nova Pro | Result |
|---|---|---|---|
| AA-SciCodeSource | 34.7% | 20.8% | Nemotron Ultra 253B leads |
Reasoning2 benchmarks
Knowledge6 benchmarks
| Benchmark | Nemotron Ultra 253B | Nova Pro | Result |
|---|---|---|---|
| Artificial Analysis Intelligence IndexSource | 9.1% | 7.7% | Nemotron Ultra 253B leads |
| AA-GPQA DiamondSource | 72.8% | 49.9% | Nemotron Ultra 253B leads |
| AA-HLESource | 8.1% | 3.4% | Nemotron Ultra 253B leads |
| AA-Omniscience IndexSource | -45.5% | -47.6% | Nemotron Ultra 253B leads |
| AA-Omniscience AccuracySource | 19.9% | 17.0% | Nemotron Ultra 253B leads |
| AA-Omniscience Hallucination RateSource | 81.7% | 77.9% | Nova Pro leads |
Multimodal1 benchmarks
| Benchmark | Nemotron Ultra 253B | Nova Pro | Result |
|---|---|---|---|
| AA-MMMU-ProSource | — | 44.3% | Not comparable |
Inst. Following1 benchmarks
| Benchmark | Nemotron Ultra 253B | Nova Pro | Result |
|---|---|---|---|
| AA-IFBenchSource | 38.2% | 38.1% | Nemotron Ultra 253B leads |
Frequently Asked Questions (3)
Can I compare Nemotron Ultra 253B and Nova Pro on BenchLM yet?
Not fully yet. BenchLM is tracking both models, but the sourced benchmark breakdown for this comparison is still coming soon.
Why does this comparison show “coming soon”?
BenchLM only shows category winners and benchmark-level calls when we have sourced results that can be compared fairly. For these models, the public benchmark coverage is not complete enough yet.
What data is available for Nemotron Ultra 253B and Nova Pro today?
Nemotron Ultra 253B: $0.00 input / $0.00 output per 1M tokens Nova Pro: Pricing unavailable Both model pages still include creator, context window, reasoning mode, and other metadata while benchmark coverage fills in.
Related Comparisons
Explore More
Choose a model with this week’s evidence
Join 2,000+ readers for ranking moves, pricing changes, and the claims that still need proof.
One email each week. Unsubscribe anytime.