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. | — | 4B |
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 | 256k |
| API pricingUSD per 1M tokens · lower wins | ||
Input priceWhat you pay for everything you send the model — your question, plus any documents or earlier conversation you include with it. | $1.25 | — |
Output priceWhat you pay for the text the model writes back. It is normally the dearer half: producing an answer costs more than reading one. | $4.25 | — |
Cached input priceA reduced rate for text you send over and over. If every request starts with the same instructions or the same document, the provider keeps a copy ready and charges less to read it again. | $0.15 | — |
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. | $1.25Meta | — |
Cheapest outputLowest output rate across every provider on OpenRouter, the lab included. May come from a different provider than the cheapest input. | $4.25Meta | — |
| $463Meta list price | — | |
| Benchmarks | ||
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% | 18.1% |
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% | — |
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. | — | 59.5% |
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% | — |
SWEAtlas CodeBase QnACodebase understanding — Questions about how an unfamiliar codebase actually works — where something is handled, what a change would touch — answered by reading the repository rather than editing it. Tests understanding rather than patch-writing. Higher is better. | 46.2% | — |
JobBenchProfessional tool use — Tests the AI on professional workplace tasks that require using real work tools — the kind of multi-step jobs an office worker handles. Higher is better. | 61.6% | — |
DeepSearchQAAgentic browsing — Questions that cannot be answered from one page: the AI has to search the web, follow the trail across several sources, and put the pieces together into an answer. Higher is better. | 85.9% | — |
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | — | 72.41% |
IFBenchInstruction following — Tests whether the AI can follow detailed instructions and satisfy multiple constraints at once. Higher is better. | — | 77.94% |
Agentic IF Index (Internal)Instruction following — Meta's internal measure of whether a model keeps following the instructions it was given while working as an agent — over a long run of tool calls, not just in a single reply. Higher is better. | 46.2% | — |
OSWorld 2.0Agentic computer use — Can the AI actually operate a computer — clicking, typing, and using real apps — to finish tasks on its own? Version 2.0 is a harder, refreshed task set. Higher is better. | 47.6% | — |
AutomationBenchBusiness workflows — Tests whether the AI can run real multi-step business workflows — the kind of end-to-end office processes companies want to automate — from start to finish. Higher is better. | 38.2% | — |
GDPval-AA v2Knowledge work — economically valuable knowledge work (v2, re-based Elo) | 1615 | — |
CharXiv Reasoning · with toolsChart reasoning — The same chart-and-figure reasoning test, run with the AI allowed to use tools — writing code to inspect the image, for instance — rather than reading the chart unaided. Scores run higher than the unaided version, so read the two as separate tests. Higher is better. | 87.6% | — |
ZeroBench (tools, pass@5)Visual reasoning — percentage of 100 visual problems solved at least once in five attempts with tools | 54% | — |
SimpleVQA (with tools)Visual reasoning — accuracy on 2,024 English and Chinese visual questions with tools | 75% | — |
MRCR · 256k-512kLong context — Tests whether the AI can find specific details buried inside a very long document, here across inputs of roughly 256k to 512k tokens — several books' worth of text. Higher is better. | 66.3% | — |
MRCR · 512k-1MLong context — The same buried-detail retrieval test run on even longer inputs, from roughly 512k up to a million tokens. Scores usually slip as the document grows, so read it against the shorter span above. Higher is better. | 55.5% | — |
| Overview | ||
| Company | Meta | NVIDIA |
| Release date | Aug 5 2026 | Oct 16 2026 |
| Access | Closed | Open Weight |
| Model details | View model | View model |
Other comparisons
Muse Spark 1.2vsClaude Fable 5.2Nemotron Labs Dory 1.0 4BvsClaude Fable 5.2Muse Spark 1.2vsGPT-6.1 TerraNemotron Labs Dory 1.0 4BvsGPT-6.1 TerraMuse Spark 1.2vsGemini 4 CarbonNemotron Labs Dory 1.0 4BvsGemini 4 CarbonMuse Spark 1.2vsGrok 4.7Nemotron Labs Dory 1.0 4BvsGrok 4.7Muse Spark 1.2vsDeepSeek-V4.1-Flash-1017Nemotron Labs Dory 1.0 4BvsDeepSeek-V4.1-Flash-1017Muse Spark 1.2vsMistral Large 4Nemotron Labs Dory 1.0 4BvsMistral Large 4Frequently asked questions
Muse Spark 1.2 leads Nemotron Labs Dory 1.0 4B on 1 of the 1 benchmark they both report (Terminal-Bench 2.1). Only Muse Spark 1.2 has a verified first-party price for an available API: $1.25 per million input tokens and $4.25 per million output tokens. No currently available pay-as-you-go API rate is tracked for Nemotron Labs Dory 1.0 4B. Muse Spark 1.2 shipped 72 days before Nemotron Labs Dory 1.0 4B, so benchmark comparisons should account for the intervening progress.
Context windows are 1M (Muse Spark 1.2) vs 256k (Nemotron Labs Dory 1.0 4B). Muse Spark 1.2 is closed, while Nemotron Labs Dory 1.0 4B is open weight.
On Terminal-Bench 2.1, Muse Spark 1.2 leads at 82.9% vs Nemotron Labs Dory 1.0 4B at 18.1%.