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. | $3.00 | $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. | $15.00 | $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. | $0.30 | $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. | $3.00Amazon 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. | $15.00Amazon Bedrock | $1.875Google AI Studio |
| $1,350Anthropic list price | $338Google 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. | 91% | 39% |
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. | 0% | 0.5% |
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. | 30% | 49% |
BU BenchBrowser agent — Can the AI drive a real web browser to finish tasks — clicking, filling forms, and navigating sites the way a person would? Run by Browser Use on their BU Bench task set. Higher is better. | 62% | 68% |
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. | 58.3% | 60.4% |
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. | 72.5% | 83% |
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. | — | 7.3% |
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. | — | 32.2% |
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. | — | 37.3% |
SWE-Bench VerifiedCoding — Real coding tasks pulled from open-source projects — the AI has to find and fix actual bugs. A human-checked version of the original SWE-Bench. Higher is better. | 79.6% | — |
MLE-BenchML engineering — Can the AI do the work of a machine-learning engineer? It competes in real Kaggle competitions — building, training, and tuning models end to end — and the score reflects how well it places. Higher is better. | — | 63.9% |
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. | 45% | — |
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. | 69.5% | — |
Humanity's Last Exam · no toolsMultidisciplinary reasoning — Humanity's Last Exam — extremely hard expert questions across many subjects, written so you can't just look up the answer. “No tools” means the AI answers on its own. Higher is better. | 33.2% | — |
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. | 49% | — |
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 89.9% | — |
Finance Agent v2Agentic financial analysis — Tests the AI on real financial-analysis work, like digging through reports and making sound decisions. Higher is better. | 51% | — |
GDPval-AAKnowledge work — Measures how well the AI does economically valuable knowledge work, judged against human experts. Shown as a rating (like a chess Elo) — higher is better. | 1676 | — |
GDPval-AA v2Knowledge work — economically valuable knowledge work (v2, re-based Elo) | — | 1421 |
CharXiv ReasoningChart reasoning — Can the AI read and reason about complex charts and figures, not just text? Higher is better. | 72.4% | — |
MMMU-ProMultimodal reasoning — A tougher version of MMMU — college-level questions that mix images, diagrams, and text together. Higher is better. | 74.5% | — |
Blueprint-Bench 2Spatial reasoning — Can the AI reason about space and layout — for example, understanding a floor plan or blueprint? Higher is better. | 6.7% | — |
MRCR v2 (8-needle) · 128k averageLong context — Tests whether the AI can find specific details buried inside a very long document (around 128k tokens — roughly a long book). Higher is better. | 84.9% | — |
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. | — | 1349 |
| Overview | ||
| Company | Anthropic | |
| Release date | Feb 17 2026 | Jul 21 2026 |
| Access | Closed | Closed |
| Model details | View model | View model |
Other comparisons
Claude Sonnet 4.6vsGPT-6.1 SolGemini 3.6 FlashvsGPT-6.1 SolClaude Sonnet 4.6vsMuse Spark OpenGemini 3.6 FlashvsMuse Spark OpenClaude Sonnet 4.6vsGrok 4.7Gemini 3.6 FlashvsGrok 4.7Claude Sonnet 4.6vsDeepSeek-V4.1-Flash-1017Gemini 3.6 FlashvsDeepSeek-V4.1-Flash-1017Claude Sonnet 4.6vsMistral Large 4Gemini 3.6 FlashvsMistral Large 4Claude Sonnet 4.6vsKimi K3Gemini 3.6 FlashvsKimi K3Frequently asked questions
Gemini 3.6 Flash leads Claude Sonnet 4.6 on 5 of the 6 benchmarks they both report. Gemini 3.6 Flash is cheaper on both input and output: $0.75 vs $3.00 per million input tokens, and $3.75 vs $15.00 per million output tokens. Claude Sonnet 4.6 shipped 154 days before Gemini 3.6 Flash, so benchmark comparisons should account for the intervening progress.
Context windows are 1M (Claude Sonnet 4.6) vs 1M (Gemini 3.6 Flash).
On BullshitBench v2, Claude Sonnet 4.6 leads at 91% vs Gemini 3.6 Flash at 39%. On ProgramBench, Gemini 3.6 Flash leads at 0.5% vs Claude Sonnet 4.6 at 0%. On DeepSWE 1.1, Gemini 3.6 Flash leads at 49% vs Claude Sonnet 4.6 at 30%. On BU Bench, Gemini 3.6 Flash leads at 68% vs Claude Sonnet 4.6 at 62%. On ARC-AGI-2, Gemini 3.6 Flash leads at 60.4% vs Claude Sonnet 4.6 at 58.3%. On OSWorld-Verified, Gemini 3.6 Flash leads at 83% vs Claude Sonnet 4.6 at 72.5%.