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 | $10.00 |
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 | $50.00 |
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 | $1.00 |
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 | $5.00OpenAI |
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 | $25.00OpenAI |
| $900Google list price | $4,500OpenAI 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% | 69% |
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% | 74.1% |
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% | 65.5% |
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. | 100% | 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. | 57.4% | 57.9% |
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% | 64.6% |
RoboDojo SimRobotics — control of a simulated two-arm robot from camera images, average success rate | 34.2% | 29.8% |
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% | 95% |
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% | 72.6% |
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% | 59.3% |
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% | 41.4% |
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% | 71% |
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 | 1889 |
CheatBenchCheating rate — Measures how often AI agents try to cheat on difficult assignments, such as reading hidden answers, copying work or manipulating grading. The overall score gives equal weight to ten categories; the sycophancy category measures how far an agent shifts its beliefs toward a user's stated views. Scores describe each model in its tested agent setup, and task success is measured separately. Lower is better. | — | 47.4% |
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% | — |
Auto-review circumvention (Internal)Safety-review circumvention — OpenAI's internal safety check on how often a model finds ways around its own automated review — the guardrail that inspects what it is about to do. This one counts failures, so lower is better and zero is the goal. | — | 0% |
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. | — | 5% |
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% | — |
FrontierCode v1.1 (Main) · main splitAgentic coding — A set of very hard, frontier-difficulty coding tasks an AI agent has to complete end to end. The score is the share of tasks in the main split it solves. Higher is better. | — | 53.3% |
FrontierCode v1.1 (Extended) · extended splitAgentic coding — frontier-difficulty agentic coding tasks (v1.1, extended split) | — | 64.5% |
FrontierCode 2.0Agentic coding — whether a model’s code changes to open-source repositories would be merged by their maintainers (version 2.0); separate from FrontierCode v1.1 scores | — | 42.8% |
AA Coding Agent IndexAgentic coding — Artificial Analysis' overall score for coding agents, combining three coding benchmarks with what each run costs and how many tokens it burns. It rates a model paired with a particular agent harness rather than the model alone, so the same model scores differently in different tools. Higher is better. | — | 67 |
Database Migration Tasks (OpenAI Internal)Database migrations — OpenAI's own test of moving a database from one schema or system to another without breaking what depends on it — the migration work that has to be right the first time. Higher is better. | — | 63.9% |
BenchCAD · with toolsCAD programming — geometric overlap of generated CAD parts with tool access | — | 95.9% |
BrowseCompWeb browsing — Can the AI browse the web and track down hard-to-find answers? Higher is better. | — | 91.5% |
ExploitBenchCybersecurity — A 'capability ladder' for security research, built by CMU researchers: the AI is given known bugs in Chrome's V8 engine and scored on how far it gets toward a working exploit inside a research sandbox — from understanding the patch to triggering a crash. Higher is better. | — | 100% |
RiemannBenchAdvanced mathematics — mathematical problem-solving evaluation published by Surge | 76% | — |
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. | — | 57.2% |
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. | — | 99.9% |
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. | — | 97.6% |
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. | — | 37.1% |
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. | — | 49.3% |
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% | — |
ARC-AGI-1Reasoning — accuracy on the first edition of the abstract grid reasoning benchmark | — | 98.5% |
LifeSciBenchReasoning — performance on research tasks graded by life sciences experts | — | 60.3% |
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | — | 96% |
CWE-bench v1Security patching — Collinear AI’s programmatically graded security-patch pass@1 rate; separate from exploit-generation benchmarks | 68% | — |
OSWorld 2.1 · offlineAgentic computer use — Can the AI actually operate a computer — clicking, typing, and using real apps — to finish tasks on its own? Version 2.1, in its offline setting. Scores are not comparable with version 2.0. Higher is better. | — | 79.5% |
SRE-BenchBinary reverse engineering — Can the model recover the behavior of compiled programs without their source code? This evaluation uses 262 binaries from 19 private programs; the score records success within four attempts. | — | 99.2% |
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 · full setHealthcare — physician rubric score on the full set of the original HealthBench evaluation | — | 56.9% |
HealthBench · full set, length-adjustedHealthcare — physician rubric score on the full set of the original HealthBench evaluation, adjusted for answer length | — | 58.3% |
HealthBench · hardHealthcare — physician rubric score on the hard subset of the original HealthBench evaluation | — | 34.2% |
HealthBench · hard, length-adjustedHealthcare — physician rubric score on the hard subset of the original HealthBench evaluation, adjusted for answer length | — | 36.6% |
HealthBench · consensusHealthcare — physician rubric score on the consensus subset of the original HealthBench evaluation | — | 95.4% |
HealthBench · consensus, length-adjustedHealthcare — physician rubric score on the consensus subset of the original HealthBench evaluation, adjusted for answer length | — | 95.5% |
HealthBench Professional · OpenAI grading, rawHealth — unadjusted physician-rubric score using the official OpenAI grading procedure | — | 68.2% |
HealthBench Professional · OpenAI grading, length-adjustedHealth — official OpenAI grading with the response-length penalty applied | — | 64.7% |
AA Intelligence IndexOverall intelligence — Artificial Analysis composite intelligence index across evals | — | 61.2 |
Design Tasks (OpenAI Internal)Design work — OpenAI's own set of professional design briefs, scored on whether the finished work is what a designer would have handed over. Higher is better. | — | 50% |
Data Science Tasks (OpenAI Internal)Data science — OpenAI's own set of data-science jobs — taking a dataset and a question through cleaning, analysis and a defensible answer. Higher is better. | — | 40.9% |
AA-Briefcase v1.1Knowledge work — Artificial Analysis agentic office-work eval (Elo, v1.1 rating fit) | — | 1569 |
AA-Briefcase v1.1 (rubric)Knowledge work — Artificial Analysis agentic office-work eval, v1.1 rubric score; separate from the Elo rating | — | 52% |
Remote Labor Index (RLI)Knowledge work — real paid remote-work projects, scored on the share completed to client standard | — | 20.8% |
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% | — |
OpenScore String QuartetsSheet music — Can the AI read sheet music? It is shown scanned pages of string quartets and has to transcribe the notation — pitches, beams, accidentals and all — with the score measuring how close the transcription lands to the real thing. Runs 0 to 1, and higher is better. | — | 0.84 |
MRCR v2 (8-needle) · 256K–512KLong context — mean match ratio for eight needles within the indicated context band | — | 100% |
MRCR v2 (8-needle) · 512K–1MLong context — mean match ratio for eight needles within the indicated context band | — | 96.3% |
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 | Sep 3 2026 |
| Access | Closed | Closed |
| Model details | View model | View model |
Other comparisons
Gemini 4 ArgonvsClaude Fable 5.2GPT-6 AstravsClaude Fable 5.2Gemini 4 ArgonvsMuse EmberGPT-6 AstravsMuse EmberGemini 4 ArgonvsGrok 4.7GPT-6 AstravsGrok 4.7Gemini 4 ArgonvsDeepSeek-V4.1-Flash-1017GPT-6 AstravsDeepSeek-V4.1-Flash-1017Gemini 4 ArgonvsMistral Large 4GPT-6 AstravsMistral Large 4Gemini 4 ArgonvsKimi K3.1 ProGPT-6 AstravsKimi K3.1 ProFrequently asked questions
GPT-6 Astra leads Gemini 4 Argon on 8 of the 13 benchmarks they both report. Gemini 4 Argon is cheaper on both input and output: $2.00 vs $10.00 per million input tokens, and $10.00 vs $50.00 per million output tokens. Figures are base-tier rates. GPT-6 Astra shipped 27 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-6 Astra).
On BullshitBench v2, GPT-6 Astra leads at 69% vs Gemini 4 Argon at 66%. On DeepSWE 1.1, Gemini 4 Argon leads at 77.9% vs GPT-6 Astra at 74.1%. On FrontierSWE V2, GPT-6 Astra leads at 65.5% vs Gemini 4 Argon at 55%. On Next.js Evals, Gemini 4 Argon leads at 100% vs GPT-6 Astra at 93%. On Terminal-Bench 4.0, GPT-6 Astra leads at 57.9% vs Gemini 4 Argon at 57.4%. On Terminal-Bench-Science 0.1, GPT-6 Astra leads at 64.6% vs Gemini 4 Argon at 57.6%. On RoboDojo Sim, Gemini 4 Argon leads at 34.2% vs GPT-6 Astra at 29.8%. On ARC-AGI-2, GPT-6 Astra leads at 95% vs Gemini 4 Argon at 91.25%. On OSWorld 2.0, GPT-6 Astra leads at 72.6% vs Gemini 4 Argon at 69.2%. On Agent's Last Exam · pass@1, GPT-6 Astra leads at 59.3% vs Gemini 4 Argon at 39.5%. On AutomationBench, Gemini 4 Argon leads at 51.3% vs GPT-6 Astra at 41.4%. On Chartography · no tools, Gemini 4 Argon leads at 71.6% vs GPT-6 Astra at 71%. On threejseval, GPT-6 Astra leads at 1889 vs Gemini 4 Argon at 1714.