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. | 816B | — |
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 | 1M |
| 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. | $1.25BaseTen | — |
Cheapest outputLowest output rate across every provider on OpenRouter, the lab included. May come from a different provider than the cheapest input. | $4.25BaseTen | — |
| $463Nebius · fp4 | — | |
| Benchmarks | ||
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. | 60% | 75.7% |
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. | 61.8% | 68.5% |
Next.js EvalsNext.js coding — Vercel's open eval of how well AI coding agents build and migrate real Next.js apps — measured as the share of tasks the agent completes successfully. Higher is better. | 93% | — |
Terminal-Bench 4.0Agentic terminal coding — Can the AI work in a command-line terminal — running commands and finishing technical setup tasks the way a developer would? Version 4.0 recalibrated how much time, CPU and memory each task gets, removed eight tasks and fixed nineteen, so fewer runs fail for reasons that have nothing to do with the model. Scores are not comparable with earlier versions. Higher is better. | — | 48.5% |
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. | 76.4% | — |
APEX-AgentsExpert agentic work — expert-level agentic work tasks (AI Productivity Index) | — | 72.1% |
MCP AtlasMulti-step tool use — Can the AI chain together many tools and steps to complete one bigger task, rather than doing just a single thing? Higher is better. | 89.2% | — |
Humanity's Last Exam · no toolsMultidisciplinary reasoning — Humanity's Last Exam — extremely hard expert questions across many subjects, written so you can't just look up the answer. “No tools” means the AI answers on its own. Higher is better. | — | 51.7% |
BioMysteryBench · human solvedBiology — Real biology puzzles that human experts have managed to crack — can the AI reach the same answers? Higher is better. | — | 85% |
Agents’ Last Exam v1.1Agentic computer use — averaged rubric score on the 122-task October 17, 2026 v1.1 set, using ALE-Claw with a five-hour limit; separate from earlier ALE scores | — | 46.7% |
AutomationBench (Artificial Analysis)Business workflows — The same business-workflow test, run by Artificial Analysis rather than by the lab, and graded strict pass@1: a workflow counts only if the whole of it is right on the first attempt. Read it as a separate board from the lab-run AutomationBench scores. Higher is better. | — | 72.5% |
HealthBench ProfessionalHealth — Realistic health conversations graded against detailed rubrics written by physicians — can the AI respond the way a careful medical professional would? Higher is better. | — | 60% |
GDPval-AA v2Knowledge work — economically valuable knowledge work (v2, re-based Elo) | 1401 | — |
AA-Briefcase v1.1Knowledge work — Artificial Analysis agentic office-work eval (Elo, v1.1 rating fit) | — | 1686 |
Chartography · no toolsChart tasks — The same chart-centred test with no tools: the AI has to read each chart unaided. Scores run far lower than the with-tools version, so read the two as separate tests. Higher is better. | — | 54.2% |
ZeroBench (tools, pass@5)Visual reasoning — percentage of 100 visual problems solved at least once in five attempts with tools | 50% | — |
SimpleVQA (with tools)Visual reasoning — accuracy on 2,024 English and Chinese visual questions with tools | 73.6% | — |
BabyVision (with tools)Visual reasoning — mean accuracy on 388 visual reasoning questions with tools | 80.2% | — |
HLE-Diamond (no tools)Reasoning — accuracy on the 1,000-question HLE-Diamond reasoning and knowledge subset without tools | 26.1% | — |
| Overview | ||
| Company | Meta | Qwen |
| Release date | Oct 15 2026 | Nov 11 2026 |
| Access | Open Weight | Closed |
| Model details | View model | View model |
Other comparisons
Muse Spark OpenvsClaude Fable 5.2Qwen4-MaxvsClaude Fable 5.2Muse Spark OpenvsGPT-6.1 TerraQwen4-MaxvsGPT-6.1 TerraMuse Spark OpenvsGemini 4 HydrogenQwen4-MaxvsGemini 4 HydrogenMuse Spark OpenvsGrok 4.7Qwen4-MaxvsGrok 4.7Muse Spark OpenvsDeepSeek-V4.1-Flash-1017Qwen4-MaxvsDeepSeek-V4.1-Flash-1017Muse Spark OpenvsMistral Large 4Qwen4-MaxvsMistral Large 4Frequently asked questions
Qwen4-Max leads Muse Spark Open on 2 of the 2 benchmarks they both report (DeepSWE 1.1, JobBench). Muse Spark Open shipped 27 days before Qwen4-Max, so benchmark comparisons should account for the intervening progress.
Context windows are 1M (Muse Spark Open) vs 1M (Qwen4-Max). Muse Spark Open is open weight, while Qwen4-Max is closed.
On DeepSWE 1.1, Qwen4-Max leads at 75.7% vs Muse Spark Open at 60%. On JobBench, Qwen4-Max leads at 68.5% vs Muse Spark Open at 61.8%.