Muse Spark 1.1vsGPT-5.6 Terra
Muse Spark 1.1 | GPT-5.6 Terra | |
|---|---|---|
| API pricingUSD per 1M tokens · lower wins · base tier | ||
Input priceWhat you pay for everything you send the model — your question, plus any documents or earlier conversation you include with it. | $1.25 | $2.00 |
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 | $12.00 |
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. | — | $0.20 |
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. | — | $2.00Azure |
Cheapest outputLowest output rate across third-party providers, excluding the lab itself. May come from a different provider than the cheapest input. | — | $12.00Azure |
| 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. | 80% | 84.3% |
Arena Elo (Text)Community preference — Real people chat with two anonymous AIs side by side and vote for the answer they prefer. Votes become a chess-style Elo rating on arena.ai — it measures which AI people actually like, not test scores. Higher is better. | 1490 | 1467 |
Arena Elo (Code)Community preference (code) — Like the text arena, but people vote on which AI writes better code. The votes become a chess-style Elo rating on arena.ai. Higher is better. | 1540 | 1521 |
| BenchmarksPublished by one model only | ||
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. | — | 53% |
Gray Swan IPI · k = 1Prompt injection robustness — Attackers hide malicious instructions inside content the AI reads — a web page, an email, a document — and try to hijack what it does. Gray Swan's indirect prompt injection benchmark measures how often such an attack succeeds when the attacker gets a single try. Lower is better. | — | 5.4% |
Gray Swan IPI · k = 10Prompt injection robustness — Attackers hide malicious instructions inside content the AI reads — a web page, an email, a document — and try to hijack what it does. This variant gives the attacker 10 tries and counts an attack as successful if any of them works. Lower is better. | — | 26% |
Gray Swan IPI · k = 15Prompt injection robustness — Attackers hide malicious instructions inside content the AI reads — a web page, an email, a document — and try to hijack what it does. This variant gives the attacker 15 tries and counts an attack as successful if any of them works. Lower is better. | — | 30.4% |
SWE-Bench ProAgentic coding — Can the AI fix real bugs in real software? It's handed actual problems from open-source projects and has to write code that genuinely solves them. Higher is better. | 61.5% | — |
CursorBench v3.2Agentic coding — Cursor's own test of harder, real-world coding tasks inside a code editor, on the refreshed v3.2 task set. Scores aren't comparable with v3.1. Higher is better. | — | 64.9% |
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. | 53.3% | — |
Frontier-Bench v0.1Agentic computer work — A hard, ever-evolving set of real computer tasks — coding, system administration, data work, and more — that an AI agent has to complete on its own. Run by the Harbor / Laude Institute team as the successor to Terminal-Bench (v0.1 is the first release of the task set). The score is the share of tasks solved. Higher is better. | — | 20.8% |
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. | 88.1% | — |
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. | 54.7% | — |
Toolathlon-VerifiedPersonal tool use — Tests how well the AI uses everyday personal tools and apps to get things done — a human-checked version of Toolathlon. Higher is better. | 75.6% | — |
Humanity's Last Exam · with toolsMultidisciplinary reasoning — Humanity's Last Exam — extremely hard expert questions across many subjects. “With tools” means the AI is allowed to search the web or run code while answering. Higher is better. | 62.1% | — |
OSWorld-VerifiedAgentic computer use — Can the AI actually operate a computer — clicking, typing, and using real apps — to finish tasks on its own? Higher is better. | 80.8% | — |
Finance Agent v2Agentic financial analysis — Tests the AI on real financial-analysis work, like digging through reports and making sound decisions. Higher is better. | 57.2% | — |
Harvey's Legal Agent BenchmarkAgentic legal work — Harvey's test of whether an AI agent can complete real legal work — drafting and reviewing documents, working with spreadsheets and presentations, and navigating files the way a lawyer's assistant would. Higher is better. | 20% | — |
TaxEval v2Tax questions — A set of real tax questions created by Vals AI — can the AI give accurate answers about tax rules and filings? Higher is better. | 79.72% | — |
MedScribeMedical admin work — Can the AI support doctors with their administrative work, like notes and paperwork? Created by Vals AI. Higher is better. | 88.89% | — |
CharXiv ReasoningChart reasoning — Can the AI read and reason about complex charts and figures, not just text? Higher is better. | 88.4% | — |
BabyVisionVisual reasoning — Tests core visual reasoning — seeing and understanding images the way even young children can, which AIs often find surprisingly hard. Higher is better. | 76.3% | — |
| Overview | ||
| Company | Meta | OpenAI |
| Release date | Jul 9 2026 | Jun 26 2026 |
| Access | Proprietary | Proprietary |
Other comparisons
Muse Spark 1.1vsClaude Opus 5GPT-5.6 TerravsClaude Opus 5Muse Spark 1.1vsGemini 3.7 FlashGPT-5.6 TerravsGemini 3.7 FlashMuse Spark 1.1vsGrok 4.6GPT-5.6 TerravsGrok 4.6Muse Spark 1.1vsDeepSeek-V4-Pro-0813GPT-5.6 TerravsDeepSeek-V4-Pro-0813Muse Spark 1.1vsMistral Medium 3.5GPT-5.6 TerravsMistral Medium 3.5Muse Spark 1.1vsKimi K3GPT-5.6 TerravsKimi K3
Frequently asked questions
Muse Spark 1.1 leads GPT-5.6 Terra on 2 of the 3 benchmarks they both report (Terminal-Bench 2.1, Arena Elo (Text), Arena Elo (Code)). Muse Spark 1.1 is cheaper on both input and output: $1.25 vs $2.00 per million input tokens, and $4.25 vs $12.00 per million output tokens. Figures are base-tier rates. GPT-5.6 Terra shipped 13 days before Muse Spark 1.1, so benchmark comparisons should account for the intervening progress.
Published specifications for these two models are limited — see each model page for the latest details.
On Terminal-Bench 2.1, GPT-5.6 Terra leads at 84.3% vs Muse Spark 1.1 at 80%. On Arena Elo (Text), Muse Spark 1.1 leads at 1490 vs GPT-5.6 Terra at 1467. On Arena Elo (Code), Muse Spark 1.1 leads at 1540 vs GPT-5.6 Terra at 1521.
Muse Spark 1.1 was released by Meta on Jul 9 2026.
GPT-5.6 Terra was released by OpenAI on Jun 26 2026.
GPT-5.6 Terra leads on Terminal-Bench 2.1 — Muse Spark 1.1 80% vs GPT-5.6 Terra 84.3%.
Muse Spark 1.1 is cheaper on both input and output: $1.25 vs $2.00 per million input tokens, and $4.25 vs $12.00 per million output tokens. Figures are base-tier rates. Rates are pay-as-you-go API prices verified on August 18, 2026.