nvidia
nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4
Hugging Face repository for Text generation.
- Context
- 262.1K
- Max output
- 65.5K
- Input / 1M
- $0.30
- Output / 1M
- $1.00
- Output speed
- —
Accepts
- Text input
Produces
- Text output
AI model comparison
Compare recorded price, context, output limits, speed, latency, capabilities, and input/output support. Every missing value stays visible, and unsourced benchmarks are excluded.
This nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4 vs trl-internal-testing/tiny-Qwen2ForCausalLM-2.5 comparison is designed for teams choosing between two production model records, not for declaring a universal winner. nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4 by nvidia currently records 262.1K context, 65.5K maximum output, $0.30 input per 1M tokens, $1.00 output per 1M tokens. trl-internal-testing/tiny-Qwen2ForCausalLM-2.5 by trl-internal-testing records limited comparable specifications. The tables below separate documented capabilities, commercial limits, directory measurements, and source-linked benchmark evidence so you can see where the data is complete and where independent testing is still required.
nvidia
Hugging Face repository for Text generation.
Accepts
Produces
trl-internal-testing
trl-internal-testing/tiny-Qwen2ForCausalLM-2.5 is a source-linked Hugging Face repository assigned to Text generation. Its recorded task interface accepts text and produces text.
Accepts
Produces
At a glance
This is a directional summary of the values recorded in this directory—not a universal quality verdict. Test the finalists on your own prompts before committing.
Where nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4 leads in recorded data
No complete recorded factor currently favors this model. That does not establish lower real-world quality.
Where trl-internal-testing/tiny-Qwen2ForCausalLM-2.5 leads in recorded data
No complete recorded factor currently favors this model. That does not establish lower real-world quality.
Context capacity
Higher listed limit
nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4:262.1K
trl-internal-testing/tiny-Qwen2ForCausalLM-2.5:—
No complete comparison
Maximum output
Higher listed limit
nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4:65.5K
trl-internal-testing/tiny-Qwen2ForCausalLM-2.5:—
No complete comparison
Input price
Lower recorded price
nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4:$0.30
trl-internal-testing/tiny-Qwen2ForCausalLM-2.5:—
No complete comparison
Output price
Lower recorded price
nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4:$1.00
trl-internal-testing/tiny-Qwen2ForCausalLM-2.5:—
No complete comparison
Output speed
Higher recorded throughput
nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4:—
trl-internal-testing/tiny-Qwen2ForCausalLM-2.5:—
No complete comparison
Time to first token
Lower recorded latency
nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4:—
trl-internal-testing/tiny-Qwen2ForCausalLM-2.5:—
No complete comparison
Intelligence Index
Higher directory index
nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4:—
trl-internal-testing/tiny-Qwen2ForCausalLM-2.5:—
No complete comparison
Evidence
A score appears only when both records include a verified source. It is not treated as a universal ranking.
Specifications
| Specification | nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4 | trl-internal-testing/tiny-Qwen2ForCausalLM-2.5 |
|---|---|---|
| Provider | nvidia | trl-internal-testing |
| Model family | — | — |
| Accepted inputs | text | text |
| Produced outputs | text | text |
| Context window | 262.1K | — |
| Max output | 65.5K | — |
| Intelligence Index | — | — |
| Input price / 1M | $0.30 | — |
| Output price / 1M | $1.00 | — |
| Output speed | — | — |
| Time to first token | — | — |
| Reasoning mode | Documented | Not documented |
| Tool calling | Documented | Not documented |
| Image input | Not documented | Not documented |
| Audio input | Not documented | Not documented |
| Structured output | Not documented | Not documented |
| License | Not documented | Not documented |
| Released | Mar 11, 2026 | Not documented |
Workload guidance
nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4 has no complete recorded factor that establishes a lead. trl-internal-testing/tiny-Qwen2ForCausalLM-2.5 has no complete recorded factor that establishes a lead. Choose according to the constraints that matter to your workload, then run the same representative prompts against both models before making a production commitment. Neither the larger number nor the lower price is automatically the better choice: prompt quality, tool reliability, modality support, rate limits, data policy, and provider implementation can change the result. Use this page to build a shortlist, confirm current terms in the linked sources, and validate quality, latency, and cost with your own traffic.
Provenance
Prices and hosted performance can change. SyncDev records source references and clearly separates documented specifications from measured values; confirm commercial terms with the provider before deployment.
nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4
nvidia · model record and recorded source
trl-internal-testing/tiny-Qwen2ForCausalLM-2.5
trl-internal-testing · model record and recorded source
Choose any two approved model records and review them with the same evidence rules.