Get started

Nemotron 3.5 Super VL

Sold by 11 sellers, timed at 214 tok/s, scored 13.6 on the Intelligence Index, and on 7 Design Arena boards.

retrieved sha256 · 0bb53a77713f

Models/NVIDIA/Nemotron 3.5 Super VLUSD per 1M tokens · as OpenRouter routes it
nvidia/nemotron-3.5-super-vlNVIDIA · base listing · Other tokenizer
OpenRouter list price · the routed rate, as the catalog carried itretrieved 17 Nov 16:29 UTC from openrouter.ai ↗
Input$0.15per 1M tokens
Output$0.50per 1M tokens
Cached input$0.02per 1M tokens
Context1.05M1,048,576 tokens
What Vitna read · 3 providers, each from its own pricing page, kept as dated recordsin / out per 1M tokens
Together AIopen weights$0.15 / $0.50confirmed 22 Nov 2026
Basetenopen weights$0.20 / $0.50confirmed 22 Nov 2026
DeepInfraopen weights$0.15 / $0.50confirmed 22 Nov 2026

The same weights served by 3 providers, each at its own list price. OpenRouter's rate is its routed rate across whichever of them it uses, and it need not match any one of them.

What the list says · the fields OpenRouter publishes for this id11 fields
AuthorNVIDIA
Listed
Canonical idnvidia/nemotron-3.5-super-vl-20261015
Context1,048,576 tokens
Max output235,929 tokens
Modalitiestext, image, video → text
TokenizerOther
Reasoningoptional
Moderationnot moderated by OpenRouter
Other listingfree variant
What the weights say · read from the model's own repository on Hugging Faceretrieved from huggingface.co ↗
Parameters124B · 124,453,395,976 declared
Model classNemotronH_Omni_Reasoning_V3
Downloadnot gated
Published

What sellers charge

11 sellers · 5x from cheapest to dearest

One tick per seller, on a log scale of input price per million from $0.06 at Darkbloom to $0.30 at Nebius. 3 sellers undercut the $0.15 listed; 3 are dearer. A faint tick declares no number format. Every seller is named on the endpoint ledger below.

The same sellers grouped by the number format each declares, on the same scale. The nvfp4 group alone spans 57% of the range, so the cheap end is not cheap for its number format. A declaration is checked by nobody: this groups the rows and certifies none.

Measured speed

13,683 requests · 11 endpoints · 30-minute window
Throughput214 tok/sMedian output rate, weighted across the 11 endpoints serving it by the requests each answered.the speed board
First token643msMedian wait before the first token, over the same window and weighted the same way.how Vitna reads a figure
Endpoint spread26 to 389The slowest and the fastest route serving this model, in tokens per second. Speed is a property of the route, not of the weights.the board, fastest first

OpenRouter's own timing of its routed traffic, weighted by requests, over one 30-minute window read . Every route is on the endpoint ledger below.

Every endpoint

11 priced · 11 timed · cheapest first
  1. Darkbloomdarkbloom/nvfp4$0.06$0.35$0.01nvfp4131k26 tok/s61.21s183 req99.04%
  2. Io Netio-net/nvfp4$0.09$0.30$0.012nvfp4131k197 tok/s53.2948ms860 req73.22%
  3. Crusoecrusoe/bf16$0.10$0.40$0.08bf16262k99 tok/s67289ms3,966 req99.92%
  4. CoreWeavecoreweave/nvfp4 · listed$0.15$0.47$0.02nvfp4944k294 tok/s112748ms3,793 req100%
  5. DeepInfradeepinfra/fp4 · listed$0.15$0.50$0.02fp4236k389 tok/s142448ms805 req99.81%
  6. PrimaLabsprimalabs/nvfp4 · listed$0.15$0.50$0.02nvfp4944k231 tok/s901.14s155 req91.66%
  7. PrimeIntellectprimeintellect/nvfp4 · listed$0.15$0.50$0.02nvfp4131k189 tok/s88789ms1,585 req99.37%
  8. Togethertogether · listed$0.15$0.50$0.02not declared236k179 tok/s53723ms1,079 req100%
  9. BaseTenbaseten/fp4$0.20$0.50$0.04fp4944k384 tok/s110492ms529 req100%
  10. Venicevenice/fp4$0.20$0.50$0.04fp4102k336 tok/s99.8925ms557 req100%
  11. Nebiusnebius/fp4$0.30$0.90$0.06fp4944k59 tok/s292.52s171 req100%

Each track runs from a route's p90, the rate nine in ten of its requests met or beat, up to its median. Weighted by requests, the figure above reports where the traffic went rather than what one route does: 6 of the 11 timed routes are slower than it, and the middle route runs at 197 tok/s. The widest median-to-p90 gap is Darkbloom's, 4.33x.

Uptime is OpenRouter's own figure for each seller over the trailing day, read once a day, and the column prints the latest reading, taken . 11 of the 12 sellers with a reading fell under 99% on at least one of the 23 days held, 23 Oct 2026 to 21 Nov 2026: Darkbloom at 25.25% on 18 Nov 2026, Venice at 51.35% on 25 Oct 2026, Io Net at 67.91% on 2 Nov 2026, Nebius (nebius/bf16) at 75.03% on 6 Nov 2026, and 7 more. Source: OpenRouter's public endpoints API, one reading a day for each seller: 1,211 sellers carried one on the latest of the 23 days held.

Source: OpenRouter's public endpoints API. 1257 sellers across 359 models when the mirror ran, 186 of them served by more than one. List prices, not what anybody paid; a discount is the seller's own. Read . Speed is OpenRouter's own timing over one 30-minute window, read .

What each seller accepts

19 parameters · 11 sellers
  1. include_reasoningevery seller
  2. max_tokensevery seller
  3. reasoningevery seller
  4. stopevery seller
  5. temperatureevery seller
  6. tool_choiceevery seller
  7. toolsevery seller
  8. top_pevery seller
  9. frequency_penalty10 of 11not PrimeIntellect
  10. presence_penalty10 of 11not PrimeIntellect
  11. response_format9 of 11not BaseTen and Io Net
  12. structured_outputs9 of 11not BaseTen and Io Net
  13. top_k9 of 11not BaseTen and PrimaLabs
  14. seed8 of 11not BaseTen, Together, and Venice
  15. logprobs7 of 11
  16. top_logprobs7 of 11
  17. repetition_penalty6 of 11
  18. logit_bias4 of 11
  19. min_p3 of 11only Crusoe, DeepInfra, and Together

8 of these 19 are accepted by every one of the 11 sellers that publish a list; the other 11 are declared by some sellers and not others, so what a request sending one of them gets depends on which seller serves it. Each list is the seller's own declaration in OpenRouter's endpoints API, read , and nothing here tested one.

Price position

342 priced models

Blended $0.238 per million tokens, 3 in to 1 out: cheaper than 260 of the 342 priced models in the catalog. Each bar is one stretch of the catalog on a logarithmic price scale; the mark is this listing.

Measured quality

145 models with an index · 132 on a design arena board

Each mark is one board this model stands on, placed at the share of that board it scores above, and led to its number in the list under the rule. The columns and the ledger below carry the figures.

Intelligence Index13.6Position 43 of 58, above 26% of the models scored.the intelligence leaderboard
Coding Indexnot scoredArtificial Analysis has not scored this model on its Coding Index.the coding leaderboard
Agentic Indexnot scoredArtificial Analysis has not scored this model on its Agentic Index.the agentic leaderboard
  1. pdf · Agents984 Elowin rate 22.6% · source rank 36the board, 36 ranked
  2. 3D scenes · Models1116 Elowin rate 30.3% · source rank 86the board, 113 ranked
  3. Code · Models1138 Elowin rate 32.3% · source rank 94the board, 120 ranked
  4. Data visualisation · Models1087 Elowin rate 26.4% · source rank 105the board, 110 ranked
  5. Game development · Models1082 Elowin rate 27.5% · source rank 107the board, 120 ranked
  6. UI components · Models1155 Elowin rate 33.1% · source rank 80the board, 116 ranked
  7. Websites · Models1157 Elowin rate 34.6% · source rank 92the board, 126 ranked
Epoch Capabilities Index · a fit by a third party over every benchmark this model has results onthe whole index, with intervals

Epoch AI has not placed Nemotron 3.5 Super VL on its Capabilities Index, or places it under a name the pairing does not reach. Epoch places a model once four of its benchmarks carry a result.

Benchmark results

none of 6 boards

None of the 6 openly licensed boards Vitna mirrors has published a result for Nemotron 3.5 Super VL under a name that matches its listing, and names are matched to the catalog exactly and never guessed. Every board, with its licence and what its publisher measures, is on /benchmarks.

Sources

Source: OpenRouter (openrouter.ai/models), as of 17 Nov 16:29 UTC. sha256 0bb53a77713f over 853,270 bytes.
Catalog: OpenRouter's public model list.
Intelligence, coding and agentic indexes: Artificial Analysis, as published by OpenRouter.
Elo, win rate and rank: Design Arena, as published by OpenRouter.
Source: each model's own Hugging Face repository (huggingface.co), as of 20 Nov 2026. 162 repositories read, 4 did not answer.
Licence, parameter count, model class and gating: each model's own Hugging Face repository, the one OpenRouter names for the listing.
A licence is the tag the uploader put on the repository, read from nothing else and verified by nobody.
A parameter count is the total the repository's safetensors files declare; for a mixture of experts that is every expert, not the share one token runs through.
Vitna ran none of these evaluations and adjusted none of these figures.