AI Tools Review

Nemotron 3.5 Super VL

By NVIDIA

Released: 2026-10-15

LLM
NVIDIA
Open Weights
Free
New

Nemotron 3.5 Super VL is NVIDIA's open-weight AI model, released 15/10/2026. NVIDIA's Hugging Face card describes a 120B-total, 12B-active hybrid Mamba-Transformer mixture-of-experts model with vision inputs, up to 1M tokens of context and the OpenMDW-1.1 licence. Results were collected through NeMo Gym against Qwen3.5-122B-A10B, Qwen3.8-Flash-Next, DeepSeek-V4-Flash0731, Nemotron 3 Ultra, Inkling-Small and GLM-5.3-Flash, and it trails the newer Qwen3.8-Flash-Next on most rows. Recommended sampling is temperature 1.0 and top_p 0.95 for text, 0.6 and 0.95 for vision; the last 11,762 characters of the page, including Ethical Considerations, were not read.

Visit Nemotron 3.5 Super VL

AI-Powered

Leverages advanced AI technology to deliver cutting-edge capabilities and results.

Fast & Efficient

Optimized performance ensures quick results without compromising on quality.

Purpose-Built

Specifically designed for llms tasks and workflows.

How does NVIDIA Nemotron 3.5 Super VL score on independent benchmarks?

NVIDIA Nemotron 3.5 Super VL ranks #28 of 28 models on the Artificial Analysis Intelligence Index (v5.0.1) with a score of 13.6, as measured on 22 November 2026.

Intelligence Index
13.6
AA v5.0.1
Coding Agent Index
n/a
Cost per task
$0.48
AA Intelligence Index task
Price per 1M tokens
$0.15 / $0.5
input / output

Where does NVIDIA Nemotron 3.5 Super VL rank against other models?

How does NVIDIA Nemotron 3.5 Super VL do on individual tests?

Evaluation (run by Artificial Analysis)NVIDIA Nemotron 3.5 Super VLBest of any model
Terminal-Bench 4.00%67.2%
FrontierCode 2.06.1%48.8%
Humanity's Last Exam30.7%74.3%
GDPval-AA (Elo)11621866

How has NVIDIA Nemotron 3.5 Super VL's score changed over time?

We have tracked NVIDIA Nemotron 3.5 Super VL on Intelligence Index v5.0.1 since 22/11/2026. A point is added each day the score, coding index or price changes, so this chart fills in over time.

What does the NVIDIA Nemotron 3.5 Super VL system card report?

NVIDIA's Hugging Face card describes a 120B-total, 12B-active hybrid Mamba-Transformer mixture-of-experts model with vision inputs, up to 1M tokens of context and the OpenMDW-1.1 licence. Results were collected through NeMo Gym against Qwen3.5-122B-A10B, Qwen3.8-Flash-Next, DeepSeek-V4-Flash0731, Nemotron 3 Ultra, Inkling-Small and GLM-5.3-Flash, and it trails the newer Qwen3.8-Flash-Next on most rows. Recommended sampling is temperature 1.0 and top_p 0.95 for text, 0.6 and 0.95 for vision; the last 11,762 characters of the page, including Ethical Considerations, were not read.

These are NVIDIA's own published results (self-reported, with the vendor's harness and settings), so they are not directly comparable with other labs' numbers. Source: nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16 on Hugging Face (15/10/2026). Last checked 22/11/2026.

Knowledge & science

BenchmarkNVIDIA Nemotron 3.5 Super VLQwen3.5-122B-A10BQwen3.8-Flash-NextDeepSeek-V4-Flash0731
GPQA-DiamondCollected via NeMo Gym unless footnoted; BF16 checkpoint89.52%86.7%92.7%89.6%
HLE (text-only)Collected via NeMo Gym unless footnoted; BF16 checkpoint31.79%24.2%39.6%35.8%
MMLU-ProXCollected via NeMo Gym unless footnoted; BF16 checkpoint76.91%n/a83%n/a
AA-Omniscience accuracyCollected via NeMo Gym unless footnoted; BF16 checkpoint26.2%n/a27.9%n/a
AA-Omniscience indexCollected via NeMo Gym unless footnoted; BF16 checkpoint4.03 pointsn/a-1.9n/a

Coding

BenchmarkNVIDIA Nemotron 3.5 Super VLQwen3.5-122B-A10BQwen3.8-Flash-NextDeepSeek-V4-Flash0731
LiveCodeBench v6Collected via NeMo Gym unless footnoted; BF16 checkpoint88.22%76.7%94.1%94.6%
SWE-bench VerifiedCollected via NeMo Gym unless footnoted; BF16 checkpoint71.1%71.9%78.7%75.4%
SWE-bench ProCollected via NeMo Gym unless footnoted; BF16 checkpoint51.78%44.9%60.37%56.91%
Terminal-Bench 2.1Collected via NeMo Gym unless footnoted; BF16 checkpoint60.67%48.7%85.8%76.8%
SciCode (subtask)Collected via NeMo Gym unless footnoted; BF16 checkpoint47.13%n/a46.6%n/a
DeepSWE (offline)Collected via NeMo Gym unless footnoted; BF16 checkpoint24.65%n/a55.57%n/a

Agents & tool use

BenchmarkNVIDIA Nemotron 3.5 Super VLQwen3.5-122B-A10BQwen3.8-Flash-NextDeepSeek-V4-Flash0731
BrowseCompCollected via NeMo Gym unless footnoted; BF16 checkpoint78%58.1%89%84.2%
tau2-bench telecomCollected via NeMo Gym unless footnoted; BF16 checkpoint73.19%87.2%90.4%98.2%
tau3-bench bankingCollected via NeMo Gym unless footnoted; BF16 checkpoint18.88%n/a44.5%n/a
AutomationBench-AA (offline)Collected via NeMo Gym unless footnoted; BF16 checkpoint32.44%n/a72.06%n/a
GDPval-AA v2Collected via NeMo Gym unless footnoted; BF16 checkpoint1211 Elon/a1561n/a
OSWorld-VerifiedCollected via NeMo Gym unless footnoted; BF16 checkpoint51.14%n/a55.8%n/a
WebVoyager (offline)Collected via NeMo Gym unless footnoted; BF16 checkpoint77.72%n/a90.39%n/a

Safety & honesty

BenchmarkNVIDIA Nemotron 3.5 Super VLQwen3.5-122B-A10BQwen3.8-Flash-NextDeepSeek-V4-Flash0731
AA-Omniscience hallucination rateLower is better; Collected via NeMo Gym unless footnoted; BF16 checkpoint30%n/a41.3%n/a

Reasoning & maths

BenchmarkNVIDIA Nemotron 3.5 Super VLQwen3.5-122B-A10BQwen3.8-Flash-NextDeepSeek-V4-Flash0731
CritPtCollected via NeMo Gym unless footnoted; BF16 checkpoint10.86%n/an/an/a
APEX shortlistCollected via NeMo Gym unless footnoted; BF16 checkpoint88.36%n/a91.8%n/a

Long context

BenchmarkNVIDIA Nemotron 3.5 Super VLQwen3.5-122B-A10BQwen3.8-Flash-NextDeepSeek-V4-Flash0731
AA-LCR v1.1Collected via NeMo Gym unless footnoted; BF16 checkpoint68.62%n/a78.4%n/a
GraphWalksCollected via NeMo Gym unless footnoted; BF16 checkpoint89.71%n/a76.1%n/a
MMLongBench-DocOfficial open-source implementation or internal scaffolding (not yet in NeMo Gym)62.82%63.7%67.9%n/a

Other

BenchmarkNVIDIA Nemotron 3.5 Super VLQwen3.5-122B-A10BQwen3.8-Flash-NextDeepSeek-V4-Flash0731
IFBenchCollected via NeMo Gym unless footnoted; BF16 checkpoint70.86%n/a75.6%n/a
WMT24++Collected via NeMo Gym unless footnoted; BF16 checkpoint84.8%n/a86.2%n/a
LMArena v3 proxyCollected via NeMo Gym unless footnoted; BF16 checkpoint52.67 pointsn/a35.2n/a
GDP.pdf (offline)Collected via NeMo Gym unless footnoted; BF16 checkpoint; card also lists an Artificial Analysis figure of 6.66.6%n/a17.4%n/a

Multimodal

BenchmarkNVIDIA Nemotron 3.5 Super VLQwen3.5-122B-A10BQwen3.8-Flash-NextDeepSeek-V4-Flash0731
V*Official open-source implementation or internal scaffolding (not yet in NeMo Gym)85.6%91.1%93%n/a
MathVisionOfficial open-source implementation or internal scaffolding (not yet in NeMo Gym)78.16%83.3%87.5%n/a
CharXiv reasoningOfficial open-source implementation or internal scaffolding (not yet in NeMo Gym)80.82%83.5%85.8%n/a
OCRBench v2 (EN)Official open-source implementation or internal scaffolding (not yet in NeMo Gym)68.12%68.3%67.9%n/a
BabyVisionOfficial open-source implementation or internal scaffolding (not yet in NeMo Gym)29.6%n/a46.8%n/a
OCR-ReasoningOfficial open-source implementation or internal scaffolding (not yet in NeMo Gym)65.08%n/a67.1%n/a
Video-MME v2Official open-source implementation or internal scaffolding (not yet in NeMo Gym)32.85%n/a45.1%n/a

What did safety testing find for NVIDIA Nemotron 3.5 Super VL?

Safety level: Not published in the readable part of the model card

  • Card mentions safety SFT and preference RL datasets in post-training, and says integration needs use-case-specific testing before deployment.

Independent figures: Artificial Analysis, refreshed daily on our AI model benchmarks page. Full analysis: our NVIDIA Nemotron 3.5 Super VL deep dive.

NVIDIA Model Timeline

Nemotron 3.5 Super VLCurrent

Up to 1M tokens (some deployment recipes use smaller windows) context

Nemotron 3 Ultra

512k tokens context

Specifications

pricing$0.15/$0.5 per 1M tokens (input/output)
context WindowUp to 1M tokens (some deployment recipes use smaller windows)