Compare AI models
| Specifications | ||
ParametersThe learned values that shape how a model responds. B means billion and T means trillion. For mixture-of-experts models, the total count is larger than the number active for each token. Parameter count alone does not measure quality. | — | 5.1B/8B/26B/31B |
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 | — |
| API pricingUSD per 1M tokens · lower wins | ||
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. | — | $0.04Darkbloom |
Cheapest outputLowest output rate across every provider on OpenRouter, the lab included. May come from a different provider than the cheapest input. | — | $0.10Makora |
| — | $14.35PrimeIntellect · bf16 | |
| Benchmarks | ||
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. | 56.8% | 19.5% |
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. | 64.7% | 26.5% |
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 94.5% | 84.3% |
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. | — | 25% |
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. | 77.8% | — |
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. | 93.9% | — |
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. | 87.3% | — |
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. | 59% | — |
LiveCodeBenchCompetitive coding — Coding problems published so recently the AI can't have seen them in training — a contamination-free test of raw programming skill. Higher is better. | — | 80% |
Terminal-Bench 2.0Agentic terminal coding — Can the AI work in a command-line terminal — running commands and finishing technical setup tasks the way a developer would? (Version 2.0 of the test.) Higher is better. | 82% | — |
CyberGymCybersecurity — Tests the AI on cybersecurity challenges — finding and exploiting software weaknesses inside a safe sandbox. Higher is better. | 83.1% | — |
AIME 2026Reasoning — single-answer accuracy on the 2026 invitational mathematics exam without tools | — | 89.2% |
MMMLUKnowledge — average accuracy on multilingual translations of MMLU | — | 88.4% |
MMLU-ProKnowledge — accuracy on the ten-option MMLU-Pro benchmark | — | 85.2% |
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. | 79.6% | — |
CharXiv ReasoningChart reasoning — Can the AI read and reason about complex charts and figures, not just text? Higher is better. | 86.1% | — |
CharXiv Reasoning · with toolsChart reasoning — The same chart-and-figure reasoning test, run with the AI allowed to use tools — writing code to inspect the image, for instance — rather than reading the chart unaided. Scores run higher than the unaided version, so read the two as separate tests. Higher is better. | 93.2% | — |
MMMU-ProMultimodal reasoning — A tougher version of MMMU — college-level questions that mix images, diagrams, and text together. Higher is better. | — | 76.9% |
GraphWalks BFS (256K–1M)Long context — breadth-first search F1 on the 200 GraphWalks problems with 256K to 1M tokens of context | 80% | — |
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. | — | 66.4% |
| Overview | ||
| Company | Anthropic | |
| Release date | Apr 7 2026 | Apr 2 2026 |
| Access | Closed | Open Weight |
| Model details | View model | View model |
Other comparisons
Claude Mythos PreviewvsGPT-6.1 TerraGemma 4vsGPT-6.1 TerraClaude Mythos PreviewvsMuse EmberGemma 4vsMuse EmberClaude Mythos PreviewvsGrok 4.7Gemma 4vsGrok 4.7Claude Mythos PreviewvsDeepSeek-V4.1-Flash-1017Gemma 4vsDeepSeek-V4.1-Flash-1017Claude Mythos PreviewvsMistral Large 4Gemma 4vsMistral Large 4Claude Mythos PreviewvsKimi K3.1 ProGemma 4vsKimi K3.1 ProFrequently asked questions
Claude Mythos Preview leads Gemma 4 on 3 of the 3 benchmarks they both report (Humanity's Last Exam, GPQA Diamond). Gemma 4 shipped 5 days before Claude Mythos Preview, so benchmark comparisons should account for the intervening progress.
Claude Mythos Preview is closed, while Gemma 4 is open weight.
On Humanity's Last Exam · no tools, Claude Mythos Preview leads at 56.8% vs Gemma 4 at 19.5%. On Humanity's Last Exam · with tools, Claude Mythos Preview leads at 64.7% vs Gemma 4 at 26.5%. On GPQA Diamond, Claude Mythos Preview leads at 94.5% vs Gemma 4 at 84.3%.