Compare AI models
| Specifications | ||
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 | 1.05M |
| 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. | $2.00 | $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. | $10.00 | $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. | $0.10 | $0.20 |
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. | $1.00Google AI Studio | $1.00OpenAI |
Cheapest outputLowest output rate across every provider on OpenRouter, the lab included. May come from a different provider than the cheapest input. | $5.00Google AI Studio | $6.00OpenAI |
| $900Google list price | $1,000OpenAI list price | |
| 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. | 66% | 54% |
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. | 0.7% | 30.4% |
CursorBench 4.0Agentic coding — Cursor's own test of coding agents on ambiguous, multi-file tasks taken from real Cursor sessions — editing, refactoring, investigating a codebase, understanding what the user meant, managing jobs and following a design. Cursor runs each model at several reasoning efforts; each release here carries the score of its best listed effort. Scores aren't comparable with earlier CursorBench versions. Higher is better. | 52.9% | 41.3% |
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. | 77.9% | 69.6% |
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. | 57.4% | 21.52% |
ARC-AGI-2Abstract reasoning — Puzzle-style tests of abstract reasoning and pattern-finding — the kind of thing people find easy but AIs often struggle with. Higher is better. | 91.25% | 83.9% |
Agent's Last Exam · pass@1Agentic computer use — A hard set of desktop and operating-system tasks an AI agent has to finish by looking at the screen and working the machine itself. The score is the share it passes outright — partial credit does not count. Higher is better. | 39.5% | 50.4% |
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. | 51.3% | 15.2% |
threejsevalCommunity preference (Three.js) — Every model gets the same prompt — "the Eiffel Tower", "a glass fishbowl", "a robot arm picking toys into a box" — and builds a 3D scene in Three.js. Real people then see two scenes side by side, names hidden, and vote for the one they prefer. The votes become a chess-style Elo rating on threejseval.com, averaged across all the prompts. It measures whether the scene looks and moves right to a human eye, not whether the code passes a test. Higher is better. | 1714 | 1429 |
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% |
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. | — | 63.4% |
FrontierSWE V2Ultra-long-horizon coding — Engineering problems that would occupy a person for days: systems implementation, performance work, scientific computing and AI research, with up to twenty hours per task. Partial credit is awarded, because finishing one outright is still rare. Higher is better. | 55% | — |
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. | 100% | — |
BenchCAD · without toolsCAD programming — geometric overlap of generated CAD parts without tool access | — | 62.3% |
BenchCAD · with toolsCAD programming — geometric overlap of generated CAD parts with tool access | — | 78.2% |
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% |
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. | — | 87.4% |
Terminal-Bench-Science 0.1Agentic scientific computing — The same command-line setup as Terminal-Bench, pointed at scientific work: the AI has to drive research tooling and computational workflows through to a result, rather than administer a machine. Version 0.1 is the first release of the task set, and scores run lower than on the general board. Higher is better. | 57.6% | — |
ToolathlonGeneral tool use — Tests how well the AI uses everyday real-world tools and apps to get things done. Higher is better. | — | 53.1% |
BrowseCompWeb browsing — Can the AI browse the web and track down hard-to-find answers? Higher is better. | — | 87.5% |
RoboDojo SimRobotics — control of a simulated two-arm robot from camera images, average success rate | 34.2% | — |
RiemannBenchAdvanced mathematics — mathematical problem-solving evaluation published by Surge | 76% | — |
ARC-AGI-3Novel problem-solving — The third generation of the ARC-AGI series: instead of static puzzles, the AI is dropped into small interactive game-like environments it has never seen and has to figure out the rules and solve them on its own. Higher is better. | — | 0.8% |
FrontierMath · Tier 4 (v2)Advanced math — The rebuilt edition of FrontierMath's hardest tier — research-level maths of the kind professional mathematicians work on. It is a different question set from the first Tier 4, and scores on it run far higher, so read the two as separate tests rather than progress. Higher is better. | — | 68.3% |
GeneBench-ProBiology — Real genomics and biomedical analyses done end to end: the AI gets a messy dataset and a question, and has to work through the chain of statistical decisions to a verifiable answer — the job a computational biologist does before a research or clinical decision gets made. Higher is better. | — | 23.3% |
MedChemBench (Internal)Chemistry — OpenAI's own drug-discovery test: reading chemical structures, predicting how potent or toxic a compound will be, choosing between candidate molecules, and planning a synthesis route — the everyday judgement calls of a medicinal chemist. Higher is better. | — | 35% |
LAB-Bench 2Biology — Everyday tasks from a working biology lab — reading protocols, interpreting figures and sequence data, and answering the practical questions a researcher hits at the bench. Higher is better. | 88.8% | — |
LifeSciBenchReasoning — performance on research tasks graded by life sciences experts | — | 56% |
FrontierMath · Tiers 1–3 (v2)Reasoning — research mathematics on the rebuilt second edition of tiers 1–3 | — | 84.9% |
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | — | 92.9% |
CWE-bench v1Security patching — Collinear AI’s programmatically graded security-patch pass@1 rate; separate from exploit-generation benchmarks | 68% | — |
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. | 69.2% | — |
Finance Agent v2Agentic financial analysis — Tests the AI on real financial-analysis work, like digging through reports and making sound decisions. Higher is better. | 65.4% | — |
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. | 19.6% | — |
HealthBench Professional · OpenAI grading, length-adjustedHealth — official OpenAI grading with the response-length penalty applied | — | 57.7% |
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. | 71.6% | — |
GDP.PDFDocument comprehension — Real professional PDFs — filings, reports, technical documents — with questions an expert in that field would ask. Tests whether the AI reads the page as a document, layout and figures included, rather than as loose text. Higher is better. | — | 24.7% |
LVBenchVideo understanding — Can the AI follow a very long video — up to an hour — and answer questions that need details from far apart in it? Higher is better. | 91.7% | — |
MMMU-Pro · with toolsMultimodal reasoning — multimodal understanding and reasoning with access to tools | — | 82% |
MMMU-ProMultimodal reasoning — A tougher version of MMMU — college-level questions that mix images, diagrams, and text together. Higher is better. | — | 80.7% |
MRCR v2 (8-needle) · 256K–512KLong context — mean match ratio for eight needles within the indicated context band | — | 89.6% |
MRCR v2 (8-needle) · 512K–1MLong context — mean match ratio for eight needles within the indicated context band | — | 72.5% |
GraphWalks BFS F1 (up to 128K)Long context — breadth-first search F1 on 650 problems with up to 128K tokens of context | 99.7% | — |
GraphWalks BFS (256K–1M)Long context — breadth-first search F1 on the 200 GraphWalks problems with 256K to 1M tokens of context | 84.2% | — |
| Overview | ||
| Company | OpenAI | |
| Release date | Sep 30 2026 | Jun 26 2026 |
| Access | Closed | Closed |
| Model details | View model | View model |
Other comparisons
Gemini 4 ArgonvsClaude Fable 5.2GPT-5.6 TerravsClaude Fable 5.2Gemini 4 ArgonvsMuse EmberGPT-5.6 TerravsMuse EmberGemini 4 ArgonvsGrok 4.7GPT-5.6 TerravsGrok 4.7Gemini 4 ArgonvsDeepSeek-V4.1-Flash-1017GPT-5.6 TerravsDeepSeek-V4.1-Flash-1017Gemini 4 ArgonvsMistral Large 4GPT-5.6 TerravsMistral Large 4Gemini 4 ArgonvsKimi K3.1 ProGPT-5.6 TerravsKimi K3.1 ProFrequently asked questions
Gemini 4 Argon leads GPT-5.6 Terra on 8 of the 9 benchmarks they both report. Both charge $2.00 per million input tokens. Gemini 4 Argon is cheaper on output: $10.00 vs $12.00 per million tokens. Figures are base-tier rates. GPT-5.6 Terra shipped 96 days before Gemini 4 Argon, so benchmark comparisons should account for the intervening progress.
Context windows are 1M (Gemini 4 Argon) vs 1.05M (GPT-5.6 Terra).
On BullshitBench v2, Gemini 4 Argon leads at 66% vs GPT-5.6 Terra at 54%. On Gray Swan IPI · k = 15, Gemini 4 Argon leads at 0.7% vs GPT-5.6 Terra at 30.4%. On CursorBench 4.0, Gemini 4 Argon leads at 52.9% vs GPT-5.6 Terra at 41.3%. On DeepSWE 1.1, Gemini 4 Argon leads at 77.9% vs GPT-5.6 Terra at 69.6%. On Terminal-Bench 4.0, Gemini 4 Argon leads at 57.4% vs GPT-5.6 Terra at 21.52%. On ARC-AGI-2, Gemini 4 Argon leads at 91.25% vs GPT-5.6 Terra at 83.9%. On Agent's Last Exam · pass@1, GPT-5.6 Terra leads at 50.4% vs Gemini 4 Argon at 39.5%. On AutomationBench, Gemini 4 Argon leads at 51.3% vs GPT-5.6 Terra at 15.2%. On threejseval, Gemini 4 Argon leads at 1714 vs GPT-5.6 Terra at 1429.