Compare AI models
| Specifications | ||
ParametersThe learned values that shape how a model responds. B means billion and T means trillion. For mixture-of-experts models, the total count is larger than the number active for each token. Parameter count alone does not measure quality. | 30B | 14B |
Context windowHow much information the model can work with in one request, measured in tokens — pieces of words or other data. This includes your question, attachments, conversation and reply. Limits can vary by model variant and provider. | 1M | 2k |
| API pricingUSD per 1M tokens · lower wins | ||
Cheapest inputLowest input rate across every provider on OpenRouter, the lab included. The cheapest endpoint may run a quantised build or a shorter context — see "Available from" on the model page. | $0.06Nebius | — |
Cheapest outputLowest output rate across every provider on OpenRouter, the lab included. May come from a different provider than the cheapest input. | $0.24Nebius | — |
| $24.00Nebius · fp8 | — | |
| These models have no shared benchmark scores. | ||
BullshitBench v2Nonsense detection — Given a confidently-worded but nonsensical prompt, does the AI spot that it makes no sense and push back — instead of playing along and inventing an answer? The score is how often it clearly called out the nonsense. Higher is better. | 28% | — |
SWE-Bench VerifiedCoding — Real coding tasks pulled from open-source projects — the AI has to find and fix actual bugs. A human-checked version of the original SWE-Bench. Higher is better. | 38.8% | — |
SWE-Bench MultilingualMultilingual coding — Like SWE-Bench, but the coding problems span many programming languages, not just one. Tests how broadly the AI can code. Higher is better. | 14.07% | — |
LiveCodeBenchCompetitive coding — Coding problems published so recently the AI can't have seen them in training — a contamination-free test of raw programming skill. Higher is better. | 68.25% | — |
HumanEvalFunction synthesis — 164 small Python problems: the AI is given a function's description and has to write the working function. This was the coding benchmark of the GPT-3.5 and GPT-4 era, before the field moved to fixing real bugs in real repositories. Higher is better. | — | 32.3% |
Terminal-Bench 2.1Agentic terminal coding — Can the AI work in a command-line terminal — running commands and finishing technical setup tasks the way a developer would? Higher is better. | 8.29% | — |
τ²-bench · retailAgentic tasks — task success in the retail domain of the second-generation customer-service evaluation | 56.91% | — |
τ²-bench · airlineAgentic tasks — task success in the airline domain of the second-generation customer-service evaluation | 48% | — |
τ²-bench · telecomAgentic tasks — task success in the telecom domain of the second-generation customer-service evaluation | 42.21% | — |
BrowseCompWeb browsing — Can the AI browse the web and track down hard-to-find answers? Higher is better. | 13.74% | — |
Humanity's Last Exam · text only, no toolsMultidisciplinary reasoning — accuracy on the 2,158 text-only questions without tools | 10.89% | — |
Humanity's Last Exam · text only, with toolsMultidisciplinary reasoning — The text-only questions from Humanity's Last Exam, answered with access to tools such as search or code execution. The score is the percentage answered correctly. | 15.48% | — |
MATHReasoning — accuracy on the original competition mathematics test set | — | 24.8% |
AIME 2025Reasoning — single-answer accuracy on the 2025 invitational mathematics exam without tools | 89.06% | — |
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 73.04% | — |
MMLU-ProKnowledge — accuracy on the ten-option MMLU-Pro benchmark | 78.3% | — |
MMLUGeneral knowledge — A 57-subject multiple-choice exam — history, law, medicine, maths — that was the standard measure of how much a model knows from 2020 until roughly 2024, when frontier scores crowded into the high 80s and labs moved on to harder tests. The scores here were published years apart under different testing setups, so read them as a historical record rather than a like-for-like ranking. Higher is better. | — | 66.3% |
GSM8KGrade-school math — Grade-school maths word problems that take a few steps of arithmetic to work through. It separated the models of 2022 and 2023 sharply, then saturated. One caveat on the historical numbers: OpenAI included part of the GSM8K training set in GPT-4's pre-training mix, so GPT-4's score is not a clean few-shot result. Higher is better. | — | 61.3% |
| Overview | ||
| Company | NVIDIA | Qwen |
| Release date | Dec 15 2025 | Sep 25 2023 |
| Access | Open Weight | Open Weight |
| Model details | View model | View model |
Other comparisons
Nemotron 3 NanovsClaude Fable 5.2Qwen-14BvsClaude Fable 5.2Nemotron 3 NanovsGPT-6.2 AstraQwen-14BvsGPT-6.2 AstraNemotron 3 NanovsGemini 4 NeonQwen-14BvsGemini 4 NeonNemotron 3 NanovsMuse EmberQwen-14BvsMuse EmberNemotron 3 NanovsGrok 4.7Qwen-14BvsGrok 4.7Nemotron 3 NanovsDeepSeek-V4.1-Flash-1017Qwen-14BvsDeepSeek-V4.1-Flash-1017Frequently asked questions
Nemotron 3 Nano and Qwen-14B don't publish scores on any of the same benchmarks, so there's no direct head-to-head comparison. Qwen-14B shipped 812 days before Nemotron 3 Nano, so benchmark comparisons should account for the intervening progress.
Nemotron 3 Nano has 30B parameters, while Qwen-14B has 14B. Context windows are 1M (Nemotron 3 Nano) vs 2k (Qwen-14B).
Direct benchmark comparisons are unavailable — Nemotron 3 Nano and Qwen-14B don't publish scores on any of the same benchmarks.