NVIDIALanguage model
Nemotron Cascade 2 30B A3B
Nemotron Cascade 2 30B A3B has no overall index value yet because there are too few independent benchmark results.
PopularityRank 366 of 785
37out of 100-1 vs. last month
based on search, usage and attention
Summary
Nemotron Cascade 2 30B A3B is a post-trained reasoning checkpoint. NVIDIA describes it as developed from a base model in the Nemotron 3 Nano line and provides separate Think and Instruct modes. Its 30 billion total parameters and 3 billion active parameters combine sparse serving with deliberate control over planning.
The model card documents tool use and supports up to 1M context tokens. NVIDIA's standard vLLM example sets the maximum model length to 262,144 tokens.
All benchmark results
Every published result with its source, test setup and comparison models.
Show all 47 published benchmark results
Page 1 of 4
Measurements without matching peer values
These measurements have no matching peer values under the same test conditions. Their original values and sources remain available here.
CritPt v2: Overall (Show measurement, test conditions, and source)
- Source value
- 0.571428571428571
- Score
- 0.571428571428571
- Metric
- pass@1 accuracy
- Unit
- %
- Benchmark version
- v2
- Category
- reasoning
- Task
- overall
- Task label
- Overall
- Direction
- Higher is better
- Source type
- independent-evaluator
- Evaluator
- Artificial Analysis
- Status
- disputed
- Retrieved at
- 2026-11-02
- Methodology
- Exact AA configuration: Nemotron Cascade 2 30B A3B (release nemotron-cascade-2-30b-a3b, row nemotron-cascade-2-30b-a3b, UUID 9815da7d-70f4-44d6-b539-9ffef0faa152, reasoning=true, effort=not specified, suffix=not specified). AA source value for average reasoning tokens before the answer: 2000. Its methodology averages 60 varied prompts and assumes 2,000 when no average is available. This is not a token cap. The full CritPt v2 run covers 70 questions with five repeats. The official grading server reports pass@1.. Source model ID: artificial-analysis/9815da7d-70f4-44d6-b539-9ffef0faa152/nemotron-cascade-2-30b-a3b/nemotron-cascade-2-30b-a3b. Source model name: Nemotron Cascade 2 30B A3B. Provider reported by source: NVIDIA. Model snapshot: Nemotron Cascade 2 30B A3B. Evaluator: Artificial Analysis. Dataset type: artificial-analysis-evaluation. Tool policy: Python functions, symbolic expressions, and numerical answers; official grading server. Attempts: 70 questions, five repeats per question, pass@1. Reasoning: reasoning enabled. Score display: 0.6%. Source ID: artificial-analysis. Evidence capture ID: aa-model-page-2026-11-02-67d46be20a5b. Source payload hash: 67d46be20a5b412db6cfb60ad190441fcaa5513500b8fd47d033168d0e97b6b2. Parser version: artificial-analysis-rsc-models-v3. Imported at: 2026-11-02. Source cohort: artificial-analysis-critpt-v2:v2:standalone:reasoning:effort=unspecified:suffix=unspecified
- Comparison methodology
- Exact AA configuration: Nemotron Cascade 2 30B A3B (release nemotron-cascade-2-30b-a3b, row nemotron-cascade-2-30b-a3b, UUID 9815da7d-70f4-44d6-b539-9ffef0faa152, reasoning=true, effort=not specified, suffix=not specified). AA source value for average reasoning tokens before the answer: 2000. Its methodology averages 60 varied prompts and assumes 2,000 when no average is available. This is not a token cap. The full CritPt v2 run covers 70 questions with five repeats. The official grading server reports pass@1.. Tool policy: Python functions, symbolic expressions, and numerical answers; official grading server. Attempts: 70 questions, five repeats per question, pass@1. Reasoning: reasoning enabled
- Context
- This AA source field contains the original CritPt score. The earlier CritPt v2 attribution was incorrect; the value and former label remain as disputed historical evidence.. The source dataset marks this result as disputed. It remains visible as documented evidence, but is not reliable comparative proof.
IFBench: Overall (Show measurement, test conditions, and source)
- Source value
- 80.4081632653061
- Score
- 80.4081632653061
- Metric
- IFBench prompt-level loose accuracy (pass@1)
- Unit
- %
- Benchmark version
- AA test set: 294 questions
- Category
- general
- Task
- overall
- Task label
- Overall
- Direction
- Higher is better
- Source type
- independent-evaluator
- Evaluator
- Artificial Analysis
- Status
- superseded
- Retrieved at
- 2026-11-02
- Methodology
- Exact AA configuration: Nemotron Cascade 2 30B A3B (release nemotron-cascade-2-30b-a3b, row nemotron-cascade-2-30b-a3b, UUID 9815da7d-70f4-44d6-b539-9ffef0faa152, reasoning=true, effort=not specified, suffix=not specified). AA source value for average reasoning tokens before the answer: 2000. Its methodology averages 60 varied prompts and assumes 2,000 when no average is available. This is not a token cap. Artificial Analysis runs 294 questions from the IFBench test set with five repeats per question. The official evaluator scores prompt-level accuracy in loose mode as pass@1.. Source model ID: artificial-analysis/9815da7d-70f4-44d6-b539-9ffef0faa152/nemotron-cascade-2-30b-a3b/nemotron-cascade-2-30b-a3b. Source model name: Nemotron Cascade 2 30B A3B. Provider reported by source: NVIDIA. Model snapshot: Nemotron Cascade 2 30B A3B. Evaluator: Artificial Analysis. Dataset type: artificial-analysis-evaluation. Tool policy: Single-turn answer evaluated with the official IFBench source code. Attempts: pass@1, 5 repeats per question. Reasoning: reasoning enabled. Score display: 80.4%. Source ID: artificial-analysis. Evidence capture ID: aa-model-page-2026-11-02-67d46be20a5b. Source payload hash: 67d46be20a5b412db6cfb60ad190441fcaa5513500b8fd47d033168d0e97b6b2. Parser version: artificial-analysis-rsc-models-v3. Imported at: 2026-11-02. Source cohort: artificial-analysis-ifbench:AA test set: 294 questions:standalone:reasoning:effort=unspecified:suffix=unspecified
- Comparison methodology
- Exact AA configuration: Nemotron Cascade 2 30B A3B (release nemotron-cascade-2-30b-a3b, row nemotron-cascade-2-30b-a3b, UUID 9815da7d-70f4-44d6-b539-9ffef0faa152, reasoning=true, effort=not specified, suffix=not specified). AA source value for average reasoning tokens before the answer: 2000. Its methodology averages 60 varied prompts and assumes 2,000 when no average is available. This is not a token cap. Artificial Analysis runs 294 questions from the IFBench test set with five repeats per question. The official evaluator scores prompt-level accuracy in loose mode as pass@1.. Tool policy: Single-turn answer evaluated with the official IFBench source code. Attempts: pass@1, 5 repeats per question. Reasoning: reasoning enabled
- Context
- The often cited 58 refers to verifiable out-of-domain constraint types, not the number of questions in AA's test run.. This result is retained as historical evidence and is not interchangeable with current results.
IFBench: Overall (Show measurement, test conditions, and source)
- Source value
- 80.4081632653061
- Score
- 80.4081632653061
- Metric
- IFBench prompt-level loose accuracy (pass@1)
- Unit
- %
- Benchmark version
- AA test set: 294 questions
- Category
- general
- Task
- overall
- Task label
- Overall
- Direction
- Higher is better
- Source type
- independent-evaluator
- Evaluator
- Artificial Analysis
- Status
- active
- Retrieved at
- 2026-11-07
- Methodology
- Exact AA configuration: Nemotron Cascade 2 30B A3B (release nemotron-cascade-2-30b-a3b, row nemotron-cascade-2-30b-a3b, UUID 9815da7d-70f4-44d6-b539-9ffef0faa152, reasoning=true, effort=not specified, suffix=not specified). This source capture does not provide an average reasoning-token count. Artificial Analysis runs 294 questions from the IFBench test set with five repeats per question. The official evaluator scores prompt-level accuracy in loose mode as pass@1.. Source model ID: artificial-analysis/9815da7d-70f4-44d6-b539-9ffef0faa152/nemotron-cascade-2-30b-a3b/nemotron-cascade-2-30b-a3b. Source model name: Nemotron Cascade 2 30B A3B. Provider reported by source: NVIDIA. Model snapshot: Nemotron Cascade 2 30B A3B. Evaluator: Artificial Analysis. Dataset type: artificial-analysis-evaluation. Tool policy: Single-turn answer evaluated with the official IFBench source code. Attempts: pass@1, 5 repeats per question. Reasoning: reasoning enabled. Score display: 80.4%. Source ID: artificial-analysis. Evidence capture ID: aa-model-page-2026-11-07-b96e3ffc8c5b. Source payload hash: b96e3ffc8c5b052aa56f96d05e89d02b8cc234a75085820fca0ca2a39e498274. Parser version: artificial-analysis-rsc-models-v4. Imported at: 2026-11-07. Source cohort: artificial-analysis-ifbench:AA test set: 294 questions:standalone:reasoning:effort=unspecified:suffix=unspecified
- Comparison methodology
- Exact AA configuration: Nemotron Cascade 2 30B A3B (release nemotron-cascade-2-30b-a3b, row nemotron-cascade-2-30b-a3b, UUID 9815da7d-70f4-44d6-b539-9ffef0faa152, reasoning=true, effort=not specified, suffix=not specified). This source capture does not provide an average reasoning-token count. Artificial Analysis runs 294 questions from the IFBench test set with five repeats per question. The official evaluator scores prompt-level accuracy in loose mode as pass@1.. Tool policy: Single-turn answer evaluated with the official IFBench source code. Attempts: pass@1, 5 repeats per question. Reasoning: reasoning enabled
- Context
- The often cited 58 refers to verifiable out-of-domain constraint types, not the number of questions in AA's test run.
HMMT February 2026 (avg@8) (Show measurement, test conditions, and source)
- Source value
- 81.06
- Score
- 81.06
- Metric
- avg@8 exact-match answer accuracy; 214 of 264 sampled answers correct
- Unit
- %
- Benchmark version
- MathArena/hmmt_feb_2026
- Category
- mathematical reasoning
- Direction
- Higher is better
- Sample
- 264
- Source type
- community-reported
- Evaluator
- Hugging Face community evaluation submission for MathArena HMMT February 2026
- Status
- active
- Retrieved at
- 2026-10-17
- Methodology
- Community-reported result. Exact checkpoint nvidia/Nemotron-Cascade-2-30B-A3B, original BF16 without quantization; vLLM 0.23.0 on 8 NVIDIA H200 GPUs with FP32 Mamba SSM cache. Thinking enabled, temperature 1.0, top-p 1.0, maximum 131,072 output tokens per response, no tools. System prompt: ‘You are a helpful and harmless assistant. You are not allowed to use any tools.’ User prompt: ‘{problem} Please place your final answer inside \boxed{}.’ Grader: math-verify==0.9.0 on the last boxed answer after </think>. Eight independently sampled responses for each of the 33 MathArena problems.
- Context
- This is a community-submitted Hugging Face result, and the PR was still open and marked Ready to merge when captured. No raw generations or confidence interval were published. The 264 observations are 33 problems × 8 responses, not 264 independent problem statements.
IOI 2025 Score@1 (Show measurement, test conditions, and source)
- Source value
- 37.2
- Score
- 37.2
- Metric
- normalized single-sample Score@1
- Unit
- %
- Benchmark version
- 2025 (arXiv:2609.02849v2)
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- NVIDIA report: Post-Training Language Models for Gold-Medal Performance in Coding Competitions (arXiv:2609.02849v2)
- Status
- active
- Retrieved at
- 2026-10-18
- Methodology
- Vendor-reported result. Nemotron-Cascade-2-30B-A3B baseline. IOI 2025 Score@1 is reported as a normalized percentage; one sampled solution is used per task per run, and the paper averages final IOI results over 1,000 runs. The NVIDIA report says results were produced with “our evaluation harness” rather than copied from the corresponding model report. The checkpoint revision, quantization, serving route, prompt, and full runtime settings are not specified for this result.
- Context
- Vendor-reported NVIDIA report, © 2026 NVIDIA. The excerpt does not establish individual authors' employment; independent evaluation is not claimed. The 1,000 repeated runs are not task N or a confidence interval. Exact checkpoint revision, quantization, route, prompt, and runtime details are not supplied.
ICPC 2025 Pass@1 (Show measurement, test conditions, and source)
- Source value
- 42
- Score
- 42
- Metric
- Pass@1, percentage of the 12 problems solved with one sampled solution
- Unit
- %
- Benchmark version
- 2025 (arXiv:2609.02849v2)
- Direction
- Higher is better
- Sample
- 12
- Source type
- vendor-reported
- Evaluator
- NVIDIA report: Post-Training Language Models for Gold-Medal Performance in Coding Competitions (arXiv:2609.02849v2)
- Status
- active
- Retrieved at
- 2026-10-18
- Methodology
- Vendor-reported result. Nemotron-Cascade-2-30B-A3B baseline. ICPC 2025 Pass@1 is the percentage of the 12 problems solved by one sampled solution; final ICPC results are averaged over 1,000 runs. The NVIDIA report says results were produced with “our evaluation harness” rather than copied from the corresponding model report. The checkpoint revision, quantization, serving route, prompt, and full runtime settings are not specified for this result.
- Context
- Vendor-reported NVIDIA report, © 2026 NVIDIA. The excerpt does not establish individual authors' employment; independent evaluation is not claimed. The 12 problems are the task denominator, separate from 1,000 repeated runs; no confidence interval is published. Exact checkpoint revision, quantization, route, prompt, and runtime details are not supplied.
LiveCodeBench Pro Pass@1 (Show measurement, test conditions, and source)
- Source value
- 45.6
- Score
- 45.6
- Metric
- Pass@1
- Unit
- %
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- NVIDIA report: Post-Training Language Models for Gold-Medal Performance in Coding Competitions (arXiv:2609.02849v2)
- Status
- active
- Retrieved at
- 2026-10-18
- Methodology
- Vendor-reported result. Nemotron-Cascade-2-30B-A3B baseline. LiveCodeBench Pro Pass@1 is the fraction of problems solved with one sampled solution; the paper averages the result over eight runs. The NVIDIA report says results were produced with “our evaluation harness” rather than copied from the corresponding model report. The LCB Pro benchmark year, snapshot, split, and item count, checkpoint revision, quantization, serving route, prompt, and full runtime settings are not specified for this result.
- Context
- Vendor-reported NVIDIA report, © 2026 NVIDIA. The excerpt does not establish individual authors' employment; independent evaluation is not claimed. The eight runs are repeated-run metadata; benchmark year, task N, snapshot, split, and confidence interval are not reported. Exact checkpoint revision, quantization, route, prompt, and runtime details are not supplied.
GDPval-AA: Overall (Show measurement, test conditions, and source)
- Source value
- 272.28
- Score
- 272.28
- Metric
- GDPval-AA Elo
- Unit
- Elo
- Benchmark version
- v2.1
- Category
- agentic
- Task
- overall
- Task label
- Overall
- Direction
- Higher is better
- Source type
- independent-evaluator
- Evaluator
- Artificial Analysis
- Status
- superseded
- Retrieved at
- 2026-11-02
- Methodology
- Exact AA configuration: Nemotron Cascade 2 30B A3B (release nemotron-cascade-2-30b-a3b, row nemotron-cascade-2-30b-a3b, UUID 9815da7d-70f4-44d6-b539-9ffef0faa152, reasoning=true, effort=not specified, suffix=not specified). AA source value for average reasoning tokens before the answer: 2000. Its methodology averages 60 varied prompts and assumes 2,000 when no average is available. This is not a token cap. 220 tasks with one run each. AA scores outcomes through pairwise panel comparisons and anchors the Elo scale at DeepSeek V4.1 Flash (Max) at 1,600.. Source model ID: artificial-analysis/9815da7d-70f4-44d6-b539-9ffef0faa152/nemotron-cascade-2-30b-a3b/nemotron-cascade-2-30b-a3b. Source model name: Nemotron Cascade 2 30B A3B. Provider reported by source: NVIDIA. Model snapshot: Nemotron Cascade 2 30B A3B. Evaluator: Artificial Analysis. Dataset type: artificial-analysis-evaluation. Harness: Artificial Analysis Stirrup. Agent: Stirrup agent. Tool policy: Agentic task completion with file outputs. Attempts: 220 tasks, one run per task; panel pairwise judgments. Reasoning: reasoning enabled. Source ID: artificial-analysis. Evidence capture ID: aa-model-page-2026-11-02-67d46be20a5b. Source payload hash: 67d46be20a5b412db6cfb60ad190441fcaa5513500b8fd47d033168d0e97b6b2. Parser version: artificial-analysis-rsc-models-v3. Imported at: 2026-11-02. Source cohort: artificial-analysis-gdpval-aa:v2.1:standalone:reasoning:effort=unspecified:suffix=unspecified
- Comparison methodology
- Exact AA configuration: Nemotron Cascade 2 30B A3B (release nemotron-cascade-2-30b-a3b, row nemotron-cascade-2-30b-a3b, UUID 9815da7d-70f4-44d6-b539-9ffef0faa152, reasoning=true, effort=not specified, suffix=not specified). AA source value for average reasoning tokens before the answer: 2000. Its methodology averages 60 varied prompts and assumes 2,000 when no average is available. This is not a token cap. 220 tasks with one run each. AA scores outcomes through pairwise panel comparisons and anchors the Elo scale at DeepSeek V4.1 Flash (Max) at 1,600.. Harness: Artificial Analysis Stirrup. Agent: Stirrup agent. Tool policy: Agentic task completion with file outputs. Attempts: 220 tasks, one run per task; panel pairwise judgments. Reasoning: reasoning enabled
- Context
- This result is retained as historical evidence and is not interchangeable with current results.
GDPval-AA: Overall (Show measurement, test conditions, and source)
- Source value
- 272.67
- Score
- 272.67
- Metric
- GDPval-AA Elo
- Unit
- Elo
- Benchmark version
- v2.1
- Category
- agentic
- Task
- overall
- Task label
- Overall
- Direction
- Higher is better
- Source type
- independent-evaluator
- Evaluator
- Artificial Analysis
- Status
- superseded
- Retrieved at
- 2026-10-29
- Methodology
- Exact AA configuration: Nemotron Cascade 2 30B A3B (release nemotron-cascade-2-30b-a3b, row nemotron-cascade-2-30b-a3b, UUID 9815da7d-70f4-44d6-b539-9ffef0faa152, reasoning=true, effort=not specified, suffix=not specified). AA source value for average reasoning tokens before the answer: 2000. Its methodology averages 60 varied prompts and assumes 2,000 when no average is available. This is not a token cap. 220 tasks with one run each. AA scores outcomes through pairwise panel comparisons and anchors the Elo scale at DeepSeek V4.1 Flash (Max) at 1,600.. Source model ID: artificial-analysis/9815da7d-70f4-44d6-b539-9ffef0faa152/nemotron-cascade-2-30b-a3b/nemotron-cascade-2-30b-a3b. Source model name: Nemotron Cascade 2 30B A3B. Provider reported by source: NVIDIA. Model snapshot: Nemotron Cascade 2 30B A3B. Evaluator: Artificial Analysis. Dataset type: artificial-analysis-evaluation. Harness: Artificial Analysis Stirrup. Agent: Stirrup agent. Tool policy: Agentic task completion with file outputs. Attempts: 220 tasks, one run per task; panel pairwise judgments. Reasoning: reasoning enabled. Source ID: artificial-analysis. Evidence capture ID: aa-model-page-2026-10-29-5a1b1cf41cd1. Source payload hash: 5a1b1cf41cd14d4fa7e1b5c737732505ff7f014395c8a19ff9385c13df09b106. Parser version: artificial-analysis-rsc-models-v3. Imported at: 2026-10-29. Source cohort: artificial-analysis-gdpval-aa:v2.1:standalone:reasoning:effort=unspecified:suffix=unspecified
- Comparison methodology
- Exact AA configuration: Nemotron Cascade 2 30B A3B (release nemotron-cascade-2-30b-a3b, row nemotron-cascade-2-30b-a3b, UUID 9815da7d-70f4-44d6-b539-9ffef0faa152, reasoning=true, effort=not specified, suffix=not specified). AA source value for average reasoning tokens before the answer: 2000. Its methodology averages 60 varied prompts and assumes 2,000 when no average is available. This is not a token cap. 220 tasks with one run each. AA scores outcomes through pairwise panel comparisons and anchors the Elo scale at DeepSeek V4.1 Flash (Max) at 1,600.. Harness: Artificial Analysis Stirrup. Agent: Stirrup agent. Tool policy: Agentic task completion with file outputs. Attempts: 220 tasks, one run per task; panel pairwise judgments. Reasoning: reasoning enabled
- Context
- This result is retained as historical evidence and is not interchangeable with current results.
GDPval-AA: Overall (Show measurement, test conditions, and source)
- Source value
- 272.28
- Score
- 272.28
- Metric
- GDPval-AA Elo
- Unit
- Elo
- Benchmark version
- v2.1
- Category
- agentic
- Task
- overall
- Task label
- Overall
- Direction
- Higher is better
- Source type
- independent-evaluator
- Evaluator
- Artificial Analysis
- Status
- active
- Retrieved at
- 2026-11-07
- Methodology
- Exact AA configuration: Nemotron Cascade 2 30B A3B (release nemotron-cascade-2-30b-a3b, row nemotron-cascade-2-30b-a3b, UUID 9815da7d-70f4-44d6-b539-9ffef0faa152, reasoning=true, effort=not specified, suffix=not specified). This source capture does not provide an average reasoning-token count. 220 tasks with one run each. AA scores outcomes through pairwise panel comparisons and anchors the Elo scale at DeepSeek V4.1 Flash (Max) at 1,600.. Source model ID: artificial-analysis/9815da7d-70f4-44d6-b539-9ffef0faa152/nemotron-cascade-2-30b-a3b/nemotron-cascade-2-30b-a3b. Source model name: Nemotron Cascade 2 30B A3B. Provider reported by source: NVIDIA. Model snapshot: Nemotron Cascade 2 30B A3B. Evaluator: Artificial Analysis. Dataset type: artificial-analysis-evaluation. Harness: Artificial Analysis Stirrup. Agent: Stirrup agent. Tool policy: Agentic task completion with file outputs. Attempts: 220 tasks, one run per task; panel pairwise judgments. Reasoning: reasoning enabled. Source ID: artificial-analysis. Evidence capture ID: aa-model-page-2026-11-07-b96e3ffc8c5b. Source payload hash: b96e3ffc8c5b052aa56f96d05e89d02b8cc234a75085820fca0ca2a39e498274. Parser version: artificial-analysis-rsc-models-v4. Imported at: 2026-11-07. Source cohort: artificial-analysis-gdpval-aa:v2.1:standalone:reasoning:effort=unspecified:suffix=unspecified
- Comparison methodology
- Exact AA configuration: Nemotron Cascade 2 30B A3B (release nemotron-cascade-2-30b-a3b, row nemotron-cascade-2-30b-a3b, UUID 9815da7d-70f4-44d6-b539-9ffef0faa152, reasoning=true, effort=not specified, suffix=not specified). This source capture does not provide an average reasoning-token count. 220 tasks with one run each. AA scores outcomes through pairwise panel comparisons and anchors the Elo scale at DeepSeek V4.1 Flash (Max) at 1,600.. Harness: Artificial Analysis Stirrup. Agent: Stirrup agent. Tool policy: Agentic task completion with file outputs. Attempts: 220 tasks, one run per task; panel pairwise judgments. Reasoning: reasoning enabled
IMO 2025 (Show measurement, test conditions, and source)
- Source value
- 35
- Score
- 35
- Metric
- points
- Unit
- points
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- NVIDIA Nemotron Cascade 2 model card
- Status
- active
- Retrieved at
- 2026-10-15
- Methodology
- Vendor-reported result. Cascade 2 column; the card explicitly labels this result as 35 points. Reasoning mode and evaluation prompt are not stated.
IMO AnswerBench (Show measurement, test conditions, and source)
- Source value
- 79.3
- Score
- 79.3
- Metric
- unknown
- Unit
- No unit provided
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- NVIDIA Nemotron Cascade 2 model card
- Status
- active
- Retrieved at
- 2026-10-15
- Methodology
- Vendor-reported result. Cascade 2 column; this is the unbracketed result, not the tool-integrated alternative. Card omits score unit and inference settings.
Page 1 of 4
Popularity compared
How often the model is searched for, used and mentioned, compared with every model of its category.
Popularity index
The 100 most popular models of this category, 0 to 100, higher is more popular
20 of 101 models
As of November 2026 · Source: gradually.ai popularity index
Over time
Weekly values from October 18, 2026 to November 7, 2026
Nemotron Cascade 2 30B A3BApertus 70B 37GLM-4.6 37Index-Translate-9B 37MiniMax M1 37
Method changes in the timeline, a jump at these points is no market event:
- 1November 1, 2026: Value only with search or usage data, 30-day attention
What is gaining popularity right now, across models, vendors, harnesses and AI subscriptions, is collected in All AI trends.
Similar models
More models of the same kind as Nemotron Cascade 2 30B A3B, chosen by model line, developer, size, price and release date.
- Nemotron 3 Nano Omni 30B A3B ReasoningNVIDIA · same developer
- Nemotron 3 NanoNVIDIA · same developer
- Nemotron 3.5 LightningNVIDIA · same developer
- Llama 3.3 Nemotron Super 49B v1.5NVIDIA · same developer
- Nemotron VoiceChatNVIDIA · same developer
- Nemotron Nano 2 VL 12BNVIDIA · same developer
- Nemotron 3.5 Bolt 74B A9BNVIDIA · same developer
- Nemotron 3 SuperNVIDIA · same developer
All specifications
Architecture
Input and output
Watermarking and provenance
AI watermarking refers to invisible signals embedded in the output. Visible marks and provenance metadata are listed separately. Unknown means there is insufficient evidence for that particular field.
- AI watermarking
- Unknown
- no source
- Watermarking technique
- Not known
- no source
Licence and access
Further details
Frequently asked questions about Nemotron Cascade 2 30B A3B
How large is the context window of Nemotron Cascade 2 30B A3B?
262,144 tokens. That is how much text Nemotron Cascade 2 30B A3B reads at once, about 390 A4 pages.
Page count as a rule of thumb from Artificial Analysis
Nemotron Cascade 2 30B A3BRank 67 of 103262K
Kimi K2.6262K
Qwen 3.8 27B262K
Kimi K2.7 Code262K
Kimi K2.5262K
Nemotron 3 Ultra262K
Kimi K2 Thinking262K
Can I run Nemotron Cascade 2 30B A3B locally?
Yes, the weights are open.
Stated by NVIDIA
- Open weights
- yes
- License
- NVIDIA Open Model License
- Parameters
- 30B
- active per token
- 3B
Sources and data date
Show all 19 sources
Benchmark sources at a glance
Which independent evaluators tested this model. This is coverage, not a quality ranking.
| Evaluator | Test series | Composite indices | Recorded active measurements |
|---|---|---|---|
| Artificial Analysis | 2 | 0 | 2 |
Without vendor claims and historical or disputed results; those stay labelled in the single values.