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. | $1.50 | $0.10 |
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. | $7.50 | $0.50 |
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. | $0.075 | $0.01 |
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. | — | $0.05OpenAI |
Cheapest outputLowest output rate across every provider on OpenRouter, the lab included. May come from a different provider than the cheapest input. | — | $0.25OpenAI |
| $675Google list price | $45.00OpenAI list price | |
| These models have no shared benchmark scores. | ||
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. | — | 59% |
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. | — | 66.6% |
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. | — | 86% |
Supabase Evals · with skillsSupabase coding — Supabase's own open benchmark: a coding agent is dropped into a real Supabase project and asked to do real work — set up a schema, fix a broken security policy, debug an Edge Function — and every run is checked against a live Supabase stack. This is the headline number, where the agent has Supabase's own skills loaded, as most people building on Supabase would. The score is the share of scenarios it got right. Higher is better. | — | 75.4% |
Supabase Evals · no skillsSupabase coding — The same Supabase scenarios, but with none of Supabase's skills loaded — so it measures what the model already knows about building on Supabase, rather than how well it follows Supabase's supplied instructions. Higher is better. | — | 63.8% |
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. | — | 16.36% |
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. | 51.4% | — |
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. | — | 59.3% |
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. | 77.6% | — |
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. | 76.1% | — |
HealthBench · full setHealthcare — physician rubric score on the full set of the original HealthBench evaluation | — | 50% |
HealthBench · full set, length-adjustedHealthcare — physician rubric score on the full set of the original HealthBench evaluation, adjusted for answer length | — | 54.5% |
HealthBench · hardHealthcare — physician rubric score on the hard subset of the original HealthBench evaluation | — | 25.4% |
HealthBench · hard, length-adjustedHealthcare — physician rubric score on the hard subset of the original HealthBench evaluation, adjusted for answer length | — | 31.4% |
HealthBench · consensusHealthcare — physician rubric score on the consensus subset of the original HealthBench evaluation | — | 95.6% |
HealthBench · consensus, length-adjustedHealthcare — physician rubric score on the consensus subset of the original HealthBench evaluation, adjusted for answer length | — | 95.9% |
HealthBench Professional · OpenAI grading, rawHealth — unadjusted physician-rubric score using the official OpenAI grading procedure | — | 61.2% |
HealthBench Professional · OpenAI grading, length-adjustedHealth — official OpenAI grading with the response-length penalty applied | — | 60.8% |
Vending-Bench 2 (USD)Business simulation — ending bank balance in US dollars in Andon Labs’ vending-business simulation; higher is better | 14549 | — |
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. | 66.6% | — |
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. | — | 1365 |
| Overview | ||
| Company | OpenAI | |
| Release date | Nov 12 2026 | Sep 22 2026 |
| Access | Closed | Closed |
| Model details | View model | View model |
Other comparisons
Gemini 4 HydrogenvsClaude Fable 5.2GPT-6 LunavsClaude Fable 5.2Gemini 4 HydrogenvsMuse EmberGPT-6 LunavsMuse EmberGemini 4 HydrogenvsGrok 4.7GPT-6 LunavsGrok 4.7Gemini 4 HydrogenvsDeepSeek-V4.1-Flash-1017GPT-6 LunavsDeepSeek-V4.1-Flash-1017Gemini 4 HydrogenvsMistral Large 4GPT-6 LunavsMistral Large 4Gemini 4 HydrogenvsKimi K3.1 ProGPT-6 LunavsKimi K3.1 ProFrequently asked questions
Gemini 4 Hydrogen and GPT-6 Luna don't publish scores on any of the same benchmarks, so there's no direct head-to-head comparison. GPT-6 Luna is cheaper on both input and output: $0.10 vs $1.50 per million input tokens, and $0.50 vs $7.50 per million output tokens. Figures are base-tier rates. GPT-6 Luna shipped 51 days before Gemini 4 Hydrogen, so benchmark comparisons should account for the intervening progress.
Context windows are 1M (Gemini 4 Hydrogen) vs 1.05M (GPT-6 Luna).
Direct benchmark comparisons are unavailable — Gemini 4 Hydrogen and GPT-6 Luna don't publish scores on any of the same benchmarks.