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. | — | 7B/14B |
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 | 1M |
| These models have no shared benchmark scores. | ||
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. | 34% | — |
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. | 72.5% | — |
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. | — | 38.6% |
HumanEvalFunction synthesis — 164 small Python problems: the AI is given a function's description and has to write the working function. This was the coding benchmark of the GPT-3.5 and GPT-4 era, before the field moved to fixing real bugs in real repositories. Higher is better. | — | 88.4% |
IFEvalAgentic tasks — following verifiable instructions in generated responses | — | 84.3% |
MATHReasoning — accuracy on the original competition mathematics test set | — | 79.5% |
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 79.6% | — |
MMLU-ProKnowledge — accuracy on the ten-option MMLU-Pro benchmark | — | 63.3% |
GSM8KGrade-school math — Grade-school maths word problems that take a few steps of arithmetic to work through. It separated the models of 2022 and 2023 sharply, then saturated. One caveat on the historical numbers: OpenAI included part of the GSM8K training set in GPT-4's pre-training mix, so GPT-4's score is not a clean few-shot result. Higher is better. | — | 94.8% |
MMMUMultimodal — Tests the AI on understanding images and text together across many college subjects. Higher is better. | 76.5% | — |
| Overview | ||
| Company | Anthropic | Qwen |
| Release date | May 22 2025 | Jan 27 2025 |
| Access | Closed | Open Weight |
| Model details | View model | View model |
Other comparisons
Claude Opus 4vsGPT-6.2 AstraQwen2.5-1MvsGPT-6.2 AstraClaude Opus 4vsGemini 4 NeonQwen2.5-1MvsGemini 4 NeonClaude Opus 4vsMuse EmberQwen2.5-1MvsMuse EmberClaude Opus 4vsGrok 4.7Qwen2.5-1MvsGrok 4.7Claude Opus 4vsDeepSeek-V4.1-Flash-1017Qwen2.5-1MvsDeepSeek-V4.1-Flash-1017Claude Opus 4vsMistral Large 4Qwen2.5-1MvsMistral Large 4Frequently asked questions
Claude Opus 4 and Qwen2.5-1M don't publish scores on any of the same benchmarks, so there's no direct head-to-head comparison. Qwen2.5-1M shipped 115 days before Claude Opus 4, so benchmark comparisons should account for the intervening progress.
Context windows are 200k (Claude Opus 4) vs 1M (Qwen2.5-1M). Claude Opus 4 is closed, while Qwen2.5-1M is open weight.
Direct benchmark comparisons are unavailable — Claude Opus 4 and Qwen2.5-1M don't publish scores on any of the same benchmarks.