Compare AI models
| Specifications | ||
ParametersA rough measure of how big the model is. More parameters usually means more capable and more expensive to run, though it is a poor guide on its own — a smaller, newer model often beats a larger, older one. | — | 120B |
Context windowHow much text the model can hold in mind at once — your question, any documents you attach, the conversation so far, and its own reply. Go past it and the earliest part falls out of view. | 1M | 1M |
| API pricingUSD per 1M tokens · lower wins | ||
Cheapest inputLowest input rate across third-party providers, excluding the lab itself. The cheapest endpoint may run a quantised build or a shorter context — see "Available from" on the model page. | $1.25Meta | $0.08DekaLLM |
Cheapest outputLowest output rate across third-party providers, excluding the lab itself. May come from a different provider than the cheapest input. | $4.25Meta | $0.40DeepInfra |
| Benchmarks | ||
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. | 50% | 54% |
ProgramBenchProgram reconstruction — The AI receives a working program and its documentation, then builds a replacement from scratch without the original source code, internet access or decompilation. The score is the percentage of 200 programs that pass every behavioral test. We record each model's best published mini-SWE-agent result, including higher reasoning efforts where available. Partial test-pass rates and almost-solved programs do not count toward this score. Equal scores share a rank here; the official board also uses partial progress to break ties. Higher is better. | 0% | — |
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. | — | 60.5% |
DeepSWE 1.1Agentic coding — Artificial Analysis' independent test of deep, agentic software-engineering work — the AI has to plan and carry out substantial coding tasks end to end. (Version 1.1 of the test.) Higher is better. | 59.3% | — |
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. | 82.9% | — |
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | — | 79.2% |
AA Intelligence IndexOverall intelligence — Artificial Analysis composite intelligence index across evals | — | 36 |
| Overview | ||
| Company | Meta | NVIDIA |
| Release date | Aug 5 2026 | Mar 11 2026 |
| Access | Closed | Open Source |
| Model details | View model | View model |
Other comparisons
Muse Spark 1.2vsClaude Haiku 5.5Nemotron 3 SupervsClaude Haiku 5.5Muse Spark 1.2vsGPT-6.1 SolNemotron 3 SupervsGPT-6.1 SolMuse Spark 1.2vsGemini 4 ArgonNemotron 3 SupervsGemini 4 ArgonMuse Spark 1.2vsGrok 4.7Nemotron 3 SupervsGrok 4.7Muse Spark 1.2vsDeepSeek-V4.1-FlashNemotron 3 SupervsDeepSeek-V4.1-FlashMuse Spark 1.2vsMistral Large 4Nemotron 3 SupervsMistral Large 4Frequently asked questions
Nemotron 3 Super leads Muse Spark 1.2 on 1 of the 1 benchmark they both report (BullshitBench v2). Nemotron 3 Super shipped 147 days before Muse Spark 1.2, so benchmark comparisons should account for the intervening progress.
Context windows are 1M (Muse Spark 1.2) vs 1M (Nemotron 3 Super). Muse Spark 1.2 is closed, while Nemotron 3 Super is open source.
On BullshitBench v2, Nemotron 3 Super leads at 54% vs Muse Spark 1.2 at 50%.