Nvidia Models

17 modelsGeneral models free to startUp to 1.05M context

Usage

Last 29 days · 2026-10-22 to 2026-11-19

Tokens

1.5B

Requests

17.1K

Models in use

14 of 17

Tokens per day, stacked by model

044.5M88.9M10-2210-2911-0511-1211-192026-10-22 — 33,459,295 tokens nemotron-3-ultra-550b-a55b-free: 19,719,045 nemotron-3.5-lightning-free: 10,554,535 nemotron-3-nano-omni-30b-a3b-reasoning-free: 3,150,825 nemotron-3.5-content-safety-free: 28,630 4 more models: 4,940 nvidia-nemotron-3-nano-30b-a3b: 660 deepinfra-nvidia-nemotron-3-nano-30b-a3b2: 6602026-10-23 — 21,163,800 tokens nemotron-3-ultra-550b-a55b-free: 14,247,625 nemotron-3.5-lightning-free: 6,381,290 nemotron-3-nano-omni-30b-a3b-reasoning-free: 499,065 nemotron-3.5-content-safety-free: 29,415 4 more models: 3,680 nvidia-nemotron-3-super-120b-a12b: 1,915 nvidia-nemotron-3-nano-30b-a3b: 405 deepinfra-nvidia-nemotron-3-nano-30b-a3b2: 4052026-10-24 — 36,540,955 tokens nemotron-3-ultra-550b-a55b-free: 28,021,365 nemotron-3-nano-omni-30b-a3b-reasoning-free: 6,098,850 nemotron-3.5-lightning-free: 2,411,510 nemotron-3.5-content-safety-free: 7,960 nvidia-nemotron-3-super-120b-a12b: 595 4 more models: 535 nvidia-nemotron-3-nano-30b-a3b: 1402026-10-25 — 41,783,095 tokens nemotron-3-ultra-550b-a55b-free: 24,556,695 nemotron-3.5-lightning-free: 12,433,375 nemotron-3.5-super-vl-free: 3,729,120 nemotron-3-nano-omni-30b-a3b-reasoning-free: 918,245 nemotron-3.5-content-safety-free: 138,440 4 more models: 4,540 nvidia-nemotron-3-super-120b-a12b: 1,270 nvidia-nemotron-3-nano-30b-a3b: 750 deepinfra-nvidia-nemotron-3-nano-30b-a3b2: 6602026-10-26 — 48,104,485 tokens nemotron-3.5-lightning-free: 22,313,620 nemotron-3-ultra-550b-a55b-free: 9,612,995 nemotron-3.5-super-vl-free: 8,152,345 nemotron-3-nano-omni-30b-a3b-reasoning-free: 7,826,850 nemotron-3.5-content-safety-free: 191,675 4 more models: 4,110 nvidia-nemotron-3-super-120b-a12b: 1,590 nvidia-nemotron-3-nano-30b-a3b: 830 deepinfra-nvidia-nemotron-3-nano-30b-a3b2: 4702026-10-27 — 39,674,010 tokens nemotron-3-ultra-550b-a55b-free: 18,382,390 nemotron-3.5-super-vl-free: 9,641,335 nemotron-3.5-lightning-free: 7,982,695 nemotron-3-nano-omni-30b-a3b-reasoning-free: 3,383,875 nemotron-3.5-content-safety-free: 188,630 4 more models: 55,375 nvidia-nemotron-3-nano-30b-a3b: 19,915 nvidia-nemotron-3-super-120b-a12b: 10,415 deepinfra-nvidia-nemotron-3-nano-30b-a3b2: 9,3802026-10-28 — 76,973,290 tokens nvidia-nemotron-3-super-120b-a12b: 43,815,565 nemotron-3-ultra-550b-a55b-free: 17,569,275 nemotron-3.5-super-vl-free: 9,742,040 nemotron-3.5-lightning-free: 5,493,630 nemotron-3-nano-omni-30b-a3b-reasoning-free: 304,170 nemotron-3.5-content-safety-free: 45,515 4 more models: 2,280 nvidia-nemotron-3-nano-30b-a3b: 540 deepinfra-nvidia-nemotron-3-nano-30b-a3b2: 2752026-10-29 — 75,938,685 tokens nemotron-3-ultra-550b-a55b-free: 60,412,725 nemotron-3.5-super-vl-free: 11,898,630 nemotron-3.5-lightning-free: 2,658,000 nemotron-3-nano-omni-30b-a3b-reasoning-free: 931,480 nemotron-3.5-content-safety-free: 36,250 4 more models: 880 nvidia-nemotron-3-super-120b-a12b: 360 nvidia-nemotron-3-nano-30b-a3b: 3602026-10-30 — 48,592,025 tokens nemotron-3.5-super-vl-free: 26,506,220 nemotron-3.5-lightning-free: 11,236,295 nemotron-3-ultra-550b-a55b-free: 10,088,025 nemotron-3-nano-omni-30b-a3b-reasoning-free: 744,385 nemotron-3.5-content-safety-free: 14,250 4 more models: 1,565 nvidia-nemotron-3-super-120b-a12b: 645 nvidia-nemotron-3-nano-30b-a3b: 500 deepinfra-nvidia-nemotron-3-nano-30b-a3b2: 1402026-10-31 — 30,843,170 tokens nemotron-3.5-super-vl-free: 16,593,435 nemotron-3-ultra-550b-a55b-free: 12,357,530 nemotron-3.5-lightning-free: 1,630,425 nemotron-3-nano-omni-30b-a3b-reasoning-free: 261,7802026-11-01 — 36,853,850 tokens nemotron-3.5-super-vl-free: 14,813,280 nemotron-3-ultra-550b-a55b-free: 13,615,500 nemotron-3.5-lightning-free: 4,818,750 nemotron-3-nano-omni-30b-a3b-reasoning-free: 3,606,3202026-11-02 — 56,741,860 tokens nemotron-3.5-super-vl-free: 27,537,825 nemotron-3-ultra-550b-a55b-free: 17,001,220 nemotron-3.5-lightning-free: 11,147,745 nemotron-3-nano-omni-30b-a3b-reasoning-free: 1,048,895 4 more models: 4,445 nvidia-nemotron-3-nano-30b-a3b: 585 nvidia-nemotron-3-super-120b-a12b: 575 deepinfra-nvidia-nemotron-3-nano-30b-a3b2: 5702026-11-03 — 86,203,230 tokens nemotron-3-ultra-550b-a55b-free: 47,021,305 nemotron-3.5-super-vl-free: 28,818,275 nemotron-3.5-lightning-free: 9,124,565 nemotron-3-nano-omni-30b-a3b-reasoning-free: 1,239,0852026-11-04 — 55,019,735 tokens nemotron-3.5-super-vl-free: 27,136,635 nemotron-3-ultra-550b-a55b-free: 22,968,410 nemotron-3.5-lightning-free: 4,508,545 nemotron-3-nano-omni-30b-a3b-reasoning-free: 404,625 4 more models: 1,250 nvidia-nemotron-3-nano-30b-a3b: 135 deepinfra-nvidia-nemotron-3-nano-30b-a3b2: 1352026-11-05 — 57,954,265 tokens nemotron-3.5-super-vl-free: 28,953,335 nemotron-3-ultra-550b-a55b-free: 25,217,655 nemotron-3-nano-omni-30b-a3b-reasoning-free: 1,959,025 nemotron-3.5-lightning-free: 1,823,630 nvidia-nemotron-3-super-120b-a12b: 265 4 more models: 240 nvidia-nemotron-3-nano-30b-a3b: 1152026-11-06 — 71,606,875 tokens nemotron-3-ultra-550b-a55b-free: 41,814,355 nemotron-3.5-super-vl-free: 28,249,545 nemotron-3.5-lightning-free: 837,455 nemotron-3-nano-omni-30b-a3b-reasoning-free: 703,330 4 more models: 1,640 nvidia-nemotron-3-super-120b-a12b: 270 nvidia-nemotron-3-nano-30b-a3b: 140 deepinfra-nvidia-nemotron-3-nano-30b-a3b2: 1402026-11-07 — 88,920,505 tokens nemotron-3-ultra-550b-a55b-free: 51,109,035 nemotron-3.5-super-vl-free: 32,633,470 nemotron-3.5-lightning-free: 4,206,425 nemotron-3-nano-omni-30b-a3b-reasoning-free: 966,425 4 more models: 3,660 nvidia-nemotron-3-nano-30b-a3b: 530 deepinfra-nvidia-nemotron-3-nano-30b-a3b2: 530 nvidia-nemotron-3-super-120b-a12b: 4302026-11-08 — 33,665,900 tokens nemotron-3.5-super-vl-free: 21,608,435 nemotron-3-ultra-550b-a55b-free: 8,896,045 nemotron-3.5-lightning-free: 3,051,380 nemotron-3-nano-omni-30b-a3b-reasoning-free: 106,575 4 more models: 2,250 nvidia-nemotron-3-super-120b-a12b: 590 deepinfra-nvidia-nemotron-3-nano-30b-a3b2: 315 nvidia-nemotron-3-nano-30b-a3b: 3102026-11-09 — 62,934,655 tokens nemotron-3.5-super-vl-free: 30,103,440 nemotron-3-ultra-550b-a55b-free: 25,113,250 nemotron-3.5-lightning-free: 5,170,575 nvidia-nemotron-3-nano-30b-a3b: 1,898,065 nemotron-3-nano-omni-30b-a3b-reasoning-free: 482,495 nvidia-nemotron-3-super-120b-a12b: 150,275 4 more models: 11,400 deepinfra-nvidia-nemotron-3-nano-30b-a3b2: 5,1552026-11-10 — 80,980,885 tokens nemotron-3.5-super-vl-free: 44,099,385 nemotron-3-ultra-550b-a55b-free: 24,740,935 nemotron-3.5-lightning-free: 7,834,725 nvidia-nemotron-3-nano-30b-a3b: 3,819,680 nemotron-3-nano-omni-30b-a3b-reasoning-free: 479,505 nvidia-nemotron-3-super-120b-a12b: 6,350 4 more models: 3052026-11-11 — 45,143,400 tokens nemotron-3.5-super-vl-free: 24,316,615 nemotron-3-ultra-550b-a55b-free: 14,070,770 nemotron-3-nano-omni-30b-a3b-reasoning-free: 3,420,630 nemotron-3.5-lightning-free: 3,323,425 nvidia-nemotron-3-super-120b-a12b: 7,820 4 more models: 3,145 nvidia-nemotron-3-nano-30b-a3b: 500 deepinfra-nvidia-nemotron-3-nano-30b-a3b2: 4952026-11-12 — 44,301,625 tokens nemotron-3-ultra-550b-a55b-free: 19,141,995 nemotron-3.5-super-vl-free: 18,921,045 nemotron-3.5-lightning-free: 5,287,265 nemotron-3-nano-omni-30b-a3b-reasoning-free: 943,580 nvidia-nemotron-3-super-120b-a12b: 6,200 4 more models: 1,130 deepinfra-nvidia-nemotron-3-nano-30b-a3b2: 275 nvidia-nemotron-3-nano-30b-a3b: 1352026-11-13 — 41,654,045 tokens nemotron-3.5-super-vl-free: 21,264,655 nemotron-3-ultra-550b-a55b-free: 12,804,255 nemotron-3.5-lightning-free: 6,339,065 nemotron-3-nano-omni-30b-a3b-reasoning-free: 1,167,835 nvidia-nemotron-3-super-120b-a12b: 71,590 4 more models: 6,140 nvidia-nemotron-3-nano-30b-a3b: 255 deepinfra-nvidia-nemotron-3-nano-30b-a3b2: 2502026-11-14 — 41,091,505 tokens nemotron-3.5-super-vl-free: 15,924,280 nemotron-3-ultra-550b-a55b-free: 15,624,015 nemotron-3.5-lightning-free: 8,055,655 nemotron-3-nano-omni-30b-a3b-reasoning-free: 1,479,050 nvidia-nemotron-3-super-120b-a12b: 7,270 4 more models: 985 nvidia-nemotron-3-nano-30b-a3b: 125 deepinfra-nvidia-nemotron-3-nano-30b-a3b2: 1252026-11-15 — 62,872,265 tokens nemotron-3.5-super-vl-free: 50,806,450 nemotron-3-ultra-550b-a55b-free: 9,478,330 nemotron-3.5-lightning-free: 2,116,695 nemotron-3-nano-omni-30b-a3b-reasoning-free: 462,290 nvidia-nemotron-3-super-120b-a12b: 7,490 4 more models: 730 nvidia-nemotron-3-nano-30b-a3b: 140 deepinfra-nvidia-nemotron-3-nano-30b-a3b2: 1402026-11-16 — 62,124,015 tokens nemotron-3-ultra-550b-a55b-free: 31,478,985 nemotron-3.5-super-vl-free: 26,069,635 nemotron-3.5-lightning-free: 4,040,990 nemotron-3-nano-omni-30b-a3b-reasoning-free: 529,165 nvidia-nemotron-3-super-120b-a12b: 5,2402026-11-17 — 26,675,145 tokens nemotron-3.5-super-vl-free: 10,210,975 nemotron-3-ultra-550b-a55b-free: 10,067,740 nemotron-3.5-lightning-free: 3,983,090 nemotron-3-nano-omni-30b-a3b-reasoning-free: 2,406,320 nvidia-nemotron-3-super-120b-a12b: 6,110 4 more models: 630 nvidia-nemotron-3-nano-30b-a3b: 140 deepinfra-nvidia-nemotron-3-nano-30b-a3b2: 1402026-11-18 — 78,907,455 tokens nemotron-3.5-super-vl-free: 39,022,030 nemotron-3-ultra-550b-a55b-free: 28,800,565 nemotron-3.5-lightning-free: 5,869,610 nemotron-3-nano-omni-30b-a3b-reasoning-free: 5,211,770 nvidia-nemotron-3-super-120b-a12b: 3,4802026-11-19 — 50,905,480 tokens nemotron-3-ultra-550b-a55b-free: 19,382,605 nemotron-3.5-super-vl-free: 15,368,140 nemotron-3.5-lightning-free: 6,074,180 nvidia-nemotron-3-super-120b-a12b: 5,369,140 nemotron-3-nano-omni-30b-a3b-reasoning-free: 4,016,015 deepinfra-nvidia-nemotron-3-nano-30b-a3b2: 372,265 4 more models: 323,135
  • nemotron-3-ultra-550b-a55b-free
  • nemotron-3.5-super-vl-free
  • nemotron-3.5-lightning-free
  • nemotron-3-nano-omni-30b-a3b-reasoning-free
  • nvidia-nemotron-3-super-120b-a12b
  • nvidia-nemotron-3-nano-30b-a3b
  • nemotron-3.5-content-safety-free
  • deepinfra-nvidia-nemotron-3-nano-30b-a3b2
  • 4 more models

Which models that traffic went to

  1. Nemotron 3 Ultra 550B A55B (free)42.5%653M
  2. Nemotron 3.5 Super VL (free)38.5%592M
  3. Nemotron 3.5 Lightning (free)11.8%181M
  4. Nemotron 3 Nano Omni 30B A3B (reasoning) (free)3.6%54.8M
  5. Nvidia Nemotron 3 Super 120B A12B3.2%49.5M
  6. Nvidia Nemotron 3 Nano 30B A3B0.4%5.7M
  7. Nemotron 3.5 Content Safety (free)<0.1%681K
  8. Deepinfra Nvidia Nemotron 3 Nano 30B A3b2<0.1%393K
  9. 4 more models<0.1%439K

Share of 1.5B tokens. 2 models with traffic report no token counts and cannot be ranked here, including nemotron-3-super-120b-a12b-free and nemotron-nano-12b-v2-vl-free — they are in the request view.

The two views disagree on purpose: a model can take a large share of the calls and a small share of the tokens — many short requests — or the reverse. Which one matters depends on whether your cost is driven by call volume or by prompt length. Measured on AIHubMix over the last 29 days, counting the 17 model IDs listed on this page; traffic routed through upstream-specific IDs that are not in the public catalog is not included.

All 17 Nvidia Models

Open in model list
Nvidia models on AIHubMix with input and output modalities, context length, maximum output, price per million tokens including cache read and cache write rates, and measured throughput and latency.
Modalities
nemotron-3-nano-30b-a3b-freeTakes text, returns text.1.05MFreeFree/M———
nemotron-3-super-120b-a12b-freeTakes text, returns text.1.05MFreeFree/M———
nemotron-3-ultra-550b-a55b-freeTakes text, returns text.1.05MFreeFree/M———
nemotron-3.5-lightning-freeTakes text, returns text.1.05MFreeFree/M———
nvidia-nemotron-3-super-120b-a12bTakes text, returns text.1M$0.11$0.55/M$0.0275/M——
nemotron-3-nano-omni-30b-a3b-reasoning-freeTakes text, vision, audio, video, returns text.262KFreeFree/M———
nemotron-3.5-content-safety-freeTakes text, vision, returns text.131KFreeFree/M———
nemotron-nano-12b-v2-vl-freeTakes text, vision, video, returns text.131KFreeFree/M———
nemotron-nano-9b-v2-freeTakes text, returns text.131KFreeFree/M———
nemotron-3.5-super-vl-freeTakes text, vision, returns text.—FreeFree/M———
nvidia-nemotron-nano-9b-v2Takes , returns text.—$0.044$0.176/M———
deepinfra-nvidia-nemotron-3-nano-30b-a3b2Takes , returns text.—$0.066$0.264/M—176 tok/s0.86 s
nvidia-nemotron-3-nano-30b-a3bTakes , returns text.—$0.066$0.264/M—176 tok/s0.86 s
nvidia-llama-3.3-nemotron-super-49b-v1.5Takes , returns text.—$0.11$0.44/M———
nvidia-nemotron-nano-12b-v2-vlTakes , returns text.—$0.22$0.66/M———
nvidia/Llama-3_1-Nemotron-Ultra-253B-v1Takes , returns text.—$0.5$0.5/M———
nvidia-llama-3.1-nemotron-70b-instructTakes , returns text.—$1.32$1.32/M———

Prices are USD per million tokens; cache read and cache write are the rates for prompt-cache hits and for writing a prompt into the cache. Throughput and latency are measured on AIHubMix — the same figures the model detail page shows — not vendor claims. A dash means the catalog does not publish that field for that model, which is not the same as the model not supporting it.

Nvidia on AIHubMix

Which Nvidia model should I start with?

nemotron-3-nano-omni-30b-a3b-reasoning-free is free on input — the cheapest entry here that declares tool calling, and it carries a 262K context. Move up to nvidia-llama-3.1-nemotron-70b-instruct when answer quality matters more than cost, or to nemotron-3-nano-30b-a3b-free for long-form reasoning.

Which of these models reason before answering?

9 of the 17 models here declare a reasoning phase — they work through the problem before producing an answer, which helps on multi-step problems at the cost of extra output tokens. Use the Reasoning filter above the table to see them. The catalog does not record anything further about how they differ, so this page does not sort them into families.

Why are there several entries for the same model?

Because each row is a route you can call, not a model release. Some IDs name an upstream (azure-, alicloud-, cc-), some are the open-weight repository form (nvidia/…), and some differ only in capitalisation, kept so older integrations keep working.

The catalog does not carry a field saying which of those a given row is, so this page does not sort them into buckets it would have to invent. Every row shows that route’s own price, context and speed — compare those directly, and open a model to see the upstreams that serve it.

How is cached input billed?

The Cache read column is the rate for input tokens served from the prompt cache — for example nvidia-nemotron-3-super-120b-a12b bills cache hits at 25% of the input rate. Cache write is the surcharge for putting a prompt into the cache in the first place, and only a few upstreams bill it separately. A dash in either column means the catalog carries no cache rate for that model, so plan on paying the full input rate.

Do I need a separate Nvidia account?

No. One AIHubMix key covers every model on this page, and switching between them is a change to the model string — billing, rate limits, and logs stay in one place.

Start calling Nvidia in one line

One key, one endpoint, 984 models across 49 model authors.