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 | 1M |
| API pricingUSD per 1M tokens · lower wins | ||
Input priceWhat you pay for everything you send the model — your question, plus any documents or earlier conversation you include with it. | $4.00 | $0.75 |
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. | $20.00 | $3.75 |
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 | $0.075 |
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. | $4.00Amazon Bedrock | $0.375Google AI Studio |
Cheapest outputLowest output rate across every provider on OpenRouter, the lab included. May come from a different provider than the cheapest input. | $20.00Amazon Bedrock | $1.875Google AI Studio |
| $1,800Anthropic list price | $338Google list price | |
| Benchmarks | ||
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. | 11.2% | 75.2% |
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. | 57.8% | 39.6% |
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. | 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. | 66.4% | 19.1% |
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. | 91.7% | 89.2% |
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. | 81.8% | 59% |
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. | 2065 | 1564 |
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. | 63% | — |
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. | 18.5% | — |
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. | — | 71% |
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. | 54.4% | — |
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. | 89.9% | — |
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. | 91.3% | — |
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. | — | 89.4% |
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. | 58.7% | — |
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. | 67.7% | — |
Humanity's Last Exam (Verified)Multidisciplinary reasoning — The re-checked edition of Humanity's Last Exam: the same extremely hard expert questions, minus the ones found to be flawed or wrongly answered. Scores on it run lower than on the original exam, so read the two as separate tests rather than a before-and-after. Higher is better. | — | 54.9% |
BioMysteryBench · hardBiology — Real unsolved-style biology puzzles — the AI has to reason its way to an answer the way a research biologist would. The “hard” split contains the toughest cases. Higher is better. | — | 56.5% |
BioMysteryBench · human solvedBiology — Real biology puzzles that human experts have managed to crack — can the AI reach the same answers? Higher is better. | — | 88.8% |
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. | — | 86.2% |
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. | 40% | — |
Finance Agent v2Agentic financial analysis — Tests the AI on real financial-analysis work, like digging through reports and making sound decisions. Higher is better. | — | 61.4% |
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. | — | 10% |
GDPval-AA v2.1Knowledge work — economically valuable knowledge work (v2.1, Crowd-BT Elo fit) | 1846 | — |
GDPval-AA v2Knowledge work — economically valuable knowledge work (v2, re-based Elo) | — | 1545 |
AA-Briefcase v1.1Knowledge work — Artificial Analysis agentic office-work eval (Elo, v1.1 rating fit) | 1822 | — |
CharXiv ReasoningChart reasoning — Can the AI read and reason about complex charts and figures, not just text? Higher is better. | — | 86.2% |
Chartography · with toolsChart tasks — A chart-centred test run with tools available to the AI, reported separately from the chart-reading benchmarks above it. Higher is better. | 89% | — |
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. | 64.4% | — |
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. | — | 35% |
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. | — | 87.1% |
LVBench · agenticVideo understanding — The same long-video test, run with the AI free to navigate the video itself — skipping around, replaying sections — rather than being shown it once. Read it as a separate test from the unaided run above. Higher is better. | — | 87.8% |
| Overview | ||
| Company | Anthropic | |
| Release date | Sep 22 2026 | Sep 2 2026 |
| Access | Closed | Closed |
| Model details | View model | View model |
Other comparisons
Claude Opus 5.5vsGPT-6.1 TerraGemini 3.8 FlashvsGPT-6.1 TerraClaude Opus 5.5vsMuse Spark OpenGemini 3.8 FlashvsMuse Spark OpenClaude Opus 5.5vsGrok 4.7Gemini 3.8 FlashvsGrok 4.7Claude Opus 5.5vsDeepSeek-V4.1-Flash-1017Gemini 3.8 FlashvsDeepSeek-V4.1-Flash-1017Claude Opus 5.5vsMistral Large 4Gemini 3.8 FlashvsMistral Large 4Claude Opus 5.5vsKimi K3.1 ProGemini 3.8 FlashvsKimi K3.1 ProFrequently asked questions
Claude Opus 5.5 leads Gemini 3.8 Flash on 7 of the 7 benchmarks they both report. Gemini 3.8 Flash is cheaper on both input and output: $0.75 vs $4.00 per million input tokens, and $3.75 vs $20.00 per million output tokens. Gemini 3.8 Flash shipped 20 days before Claude Opus 5.5, so benchmark comparisons should account for the intervening progress.
Context windows are 1M (Claude Opus 5.5) vs 1M (Gemini 3.8 Flash).
On CheatBench, Claude Opus 5.5 leads at 11.2% vs Gemini 3.8 Flash at 75.2%. On CursorBench 4.0, Claude Opus 5.5 leads at 57.8% vs Gemini 3.8 Flash at 39.6%. On Next.js Evals, Claude Opus 5.5 leads at 100% vs Gemini 3.8 Flash at 93%. On Terminal-Bench 4.0, Claude Opus 5.5 leads at 66.4% vs Gemini 3.8 Flash at 19.1%. On ARC-AGI-2, Claude Opus 5.5 leads at 91.7% vs Gemini 3.8 Flash at 89.2%. On OSWorld 2.0, Claude Opus 5.5 leads at 81.8% vs Gemini 3.8 Flash at 59%. On threejseval, Claude Opus 5.5 leads at 2065 vs Gemini 3.8 Flash at 1564.