Compare AI models
| Specifications | ||
ParametersA rough measure of how big the model is. More parameters usually means more capable and more expensive to run, though it is a poor guide on its own — a smaller, newer model often beats a larger, older one. | — | 2.8T |
Context windowHow much text the model can hold in mind at once — your question, any documents you attach, the conversation so far, and its own reply. Go past it and the earliest part falls out of view. | 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. | $10.00 | $3.00 |
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. | $50.00 | $15.00 |
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. | $0.25 | $0.30 |
Cheapest inputLowest input rate across third-party providers, excluding the lab itself. The cheapest endpoint may run a quantised build or a shorter context — see "Available from" on the model page. | $10.00Amazon Bedrock | $0.83Relace |
Cheapest outputLowest output rate across third-party providers, excluding the lab itself. May come from a different provider than the cheapest input. | $50.00Amazon Bedrock | $9.75Phala |
| 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. | 45.8% | 70% |
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. | 77% | 74% |
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. | 97% | 84% |
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. | 60.9% | 43.5% |
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. | 65.6% | 56% |
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. | 90% | 60.4% |
GDPval-AA v2Knowledge work — economically valuable knowledge work (v2, re-based Elo) | 1853 | 1668 |
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. | 1849 | 1508 |
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. | 81.2% | — |
SWE-Bench MultilingualMultilingual coding — Like SWE-Bench, but the coding problems span many programming languages, not just one. Tests how broadly the AI can code. Higher is better. | 89.1% | — |
SWE-Bench MultimodalMultimodal coding — Real bug reports that arrive with pictures attached — a screenshot, a mockup, a page rendering wrongly — so the AI has to read the image as well as the code to work out what to fix. Higher is better. | 54.7% | — |
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. | 51.8% | — |
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. | — | 69% |
DeepSWE 1.0Agentic 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. Higher is better. | — | 67.5% |
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. | 50.3% | — |
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. | — | 87% |
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. | — | 84.1% |
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. | 57.88% | — |
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. | — | 88.3% |
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. | 52.6% | — |
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. | — | 84.2% |
JobBenchProfessional tool use — Tests the AI on professional workplace tasks that require using real work tools — the kind of multi-step jobs an office worker handles. Higher is better. | — | 52.9% |
Toolathlon-VerifiedPersonal tool use — Tests how well the AI uses everyday personal tools and apps to get things done — a human-checked version of Toolathlon. Higher is better. | — | 73.2% |
BrowseCompWeb browsing — Can the AI browse the web and track down hard-to-find answers? Higher is better. | — | 91.2% |
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | — | 93.5% |
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. | 80.7% | — |
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. | 31.4% | — |
HealthBench ProfessionalHealth — Realistic health conversations graded against detailed rubrics written by physicians — can the AI respond the way a careful medical professional would? Higher is better. | 62.1% | — |
GDPval-AA v2.1Knowledge work — economically valuable knowledge work (v2.1, Crowd-BT Elo fit) | 1735 | — |
AA-BriefcaseKnowledge work — Artificial Analysis agentic office-work eval (Elo) | 1694 | — |
CharXiv ReasoningChart reasoning — Can the AI read and reason about complex charts and figures, not just text? Higher is better. | — | 84.8% |
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. | 88.4% | — |
MMMU-ProMultimodal reasoning — A tougher version of MMMU — college-level questions that mix images, diagrams, and text together. Higher is better. | — | 81.6% |
| Overview | ||
| Company | Anthropic | Moonshot AI |
| Release date | Sep 1 2026 | Jul 16 2026 |
| Access | Closed | Open Weight |
| Model details | View model | View model |
Other comparisons
Claude Fable 5.1vsGPT-6.1 SolKimi K3vsGPT-6.1 SolClaude Fable 5.1vsGemini 4 ArgonKimi K3vsGemini 4 ArgonClaude Fable 5.1vsMuse Spark 1.3Kimi K3vsMuse Spark 1.3Claude Fable 5.1vsGrok 4.7Kimi K3vsGrok 4.7Claude Fable 5.1vsDeepSeek-V4.1-FlashKimi K3vsDeepSeek-V4.1-FlashClaude Fable 5.1vsMistral Large 4Kimi K3vsMistral Large 4Frequently asked questions
Claude Fable 5.1 leads Kimi K3 on 8 of the 8 benchmarks they both report. Kimi K3 is cheaper on both input and output: $3.00 vs $10.00 per million input tokens, and $15.00 vs $50.00 per million output tokens. Kimi K3 shipped 47 days before Claude Fable 5.1, so benchmark comparisons should account for the intervening progress.
Context windows are 1M (Claude Fable 5.1) vs 1M (Kimi K3). Claude Fable 5.1 is closed, while Kimi K3 is open weight.
On CheatBench, Claude Fable 5.1 leads at 45.8% vs Kimi K3 at 70%. On BullshitBench v2, Claude Fable 5.1 leads at 77% vs Kimi K3 at 74%. On Next.js Evals, Claude Fable 5.1 leads at 97% vs Kimi K3 at 84%. On Humanity's Last Exam · no tools, Claude Fable 5.1 leads at 60.9% vs Kimi K3 at 43.5%. On Humanity's Last Exam · with tools, Claude Fable 5.1 leads at 65.6% vs Kimi K3 at 56%. On ARC-AGI-2, Claude Fable 5.1 leads at 90% vs Kimi K3 at 60.4%. On GDPval-AA v2, Claude Fable 5.1 leads at 1853 vs Kimi K3 at 1668. On threejseval, Claude Fable 5.1 leads at 1849 vs Kimi K3 at 1508.