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. | 200k | 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. | $1.00 | $1.00 |
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. | $5.00 | $5.00 |
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 | $0.05 |
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. | $1.00Amazon Bedrock | — |
Cheapest outputLowest output rate across every provider on OpenRouter, the lab included. May come from a different provider than the cheapest input. | $5.00Amazon Bedrock | — |
| $450Anthropic list price | $450OpenAI list price | |
| Benchmarks | ||
OSWorld 2.1 (offline)Agentic 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. | 15.7% | 74.2% |
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% | — |
OpenAI Factuality (Internal)Factual error rate — OpenAI's internal test uses prompts designed to provoke factual mistakes. The score is the percentage of answers containing at least one factual error, at the model's best published reasoning effort. It is not the percentage of individual claims that are wrong or an error rate for everyday use. Lower is better. | — | 4.3% |
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% | — |
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. | 73.3% | — |
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. | — | 50.3% |
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. | — | 58.3% |
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 73% | — |
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. | — | 60% |
MMMU-ProMultimodal reasoning — A tougher version of MMMU — college-level questions that mix images, diagrams, and text together. Higher is better. | — | 85.5% |
| Overview | ||
| Company | Anthropic | OpenAI |
| Release date | Oct 15 2025 | Oct 26 2026 |
| Access | Closed | Closed |
| Model details | View model | View model |
Other comparisons
Claude Haiku 4.5vsGemini 4 ArgonGPT-6.1 TerravsGemini 4 ArgonClaude Haiku 4.5vsMuse Spark OpenGPT-6.1 TerravsMuse Spark OpenClaude Haiku 4.5vsGrok 4.7GPT-6.1 TerravsGrok 4.7Claude Haiku 4.5vsDeepSeek-V4.1-Flash-1017GPT-6.1 TerravsDeepSeek-V4.1-Flash-1017Claude Haiku 4.5vsMistral Large 4GPT-6.1 TerravsMistral Large 4Claude Haiku 4.5vsKimi K3.1 ProGPT-6.1 TerravsKimi K3.1 ProFrequently asked questions
GPT-6.1 Terra leads Claude Haiku 4.5 on 1 of the 1 benchmark they both report (OSWorld 2.1 (offline)). Both charge $1.00 per million input tokens. Both charge $5.00 per million output tokens. Figures are base-tier rates. Claude Haiku 4.5 shipped 376 days before GPT-6.1 Terra, so benchmark comparisons should account for the intervening progress.
Context windows are 200k (Claude Haiku 4.5) vs 1.05M (GPT-6.1 Terra).
On OSWorld 2.1 (offline), GPT-6.1 Terra leads at 74.2% vs Claude Haiku 4.5 at 15.7%.