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. | — | 6B |
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 | 8k |
| 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. | $15.00Amazon Bedrock | — |
Cheapest outputLowest output rate across every provider on OpenRouter, the lab included. May come from a different provider than the cheapest input. | $75.00Amazon Bedrock | — |
| $6,750Amazon Bedrock | — | |
| 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. | 43% | — |
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. | 74.5% | — |
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. | — | 35.9% |
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. | 46.5% | — |
MCP Atlas · original harnessMulti-step tool use — pass rate under the original tool-use harness, before Scale changed the judge and retry handling in April 2026 | 40.9% | — |
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 80.9% | — |
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. | 44.4% | — |
MMMUMultimodal — Tests the AI on understanding images and text together across many college subjects. Higher is better. | 77.1% | — |
| Overview | ||
| Company | Anthropic | Z.ai |
| Release date | Aug 5 2025 | Jul 25 2023 |
| Access | Closed | Open Weight |
| Model details | View model | View model |
Other comparisons
Claude Opus 4.1vsGPT-6.2 AstraCodeGeeX2-6BvsGPT-6.2 AstraClaude Opus 4.1vsGemini 4 HydrogenCodeGeeX2-6BvsGemini 4 HydrogenClaude Opus 4.1vsMuse EmberCodeGeeX2-6BvsMuse EmberClaude Opus 4.1vsGrok 4.7CodeGeeX2-6BvsGrok 4.7Claude Opus 4.1vsDeepSeek-V4.1-Flash-1017CodeGeeX2-6BvsDeepSeek-V4.1-Flash-1017Claude Opus 4.1vsMistral Large 4CodeGeeX2-6BvsMistral Large 4Frequently asked questions
Claude Opus 4.1 and CodeGeeX2-6B don't publish scores on any of the same benchmarks, so there's no direct head-to-head comparison. CodeGeeX2-6B shipped 742 days before Claude Opus 4.1, so benchmark comparisons should account for the intervening progress.
Context windows are 200k (Claude Opus 4.1) vs 8k (CodeGeeX2-6B). Claude Opus 4.1 is closed, while CodeGeeX2-6B is open weight.
Direct benchmark comparisons are unavailable — Claude Opus 4.1 and CodeGeeX2-6B don't publish scores on any of the same benchmarks.