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. | — | 27B/35B/122B |
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. | 400k | 256k |
| 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. | $0.25 | — |
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. | $2.00 | — |
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. | $0.025 | — |
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. | $0.125OpenAI | — |
Cheapest outputLowest output rate across every provider on OpenRouter, the lab included. May come from a different provider than the cheapest input. | $1.00OpenAI | — |
| $150OpenAI list price | — | |
| Benchmarks | ||
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. | 71% | 72% |
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. | 16.7% | 25.3% |
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 82.3% | 86.6% |
MMMUMultimodal — Tests the AI on understanding images and text together across many college subjects. Higher is better. | 81.6% | 83.9% |
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% | — |
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. | — | 78.9% |
Aider Polyglot (diff)Coding — editing code in six languages using the diff edit format | 71.6% | — |
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. | — | 49.4% |
IFEvalAgentic tasks — following verifiable instructions in generated responses | — | 93.4% |
τ²-bench · retailAgentic tasks — task success in the retail domain of the second-generation customer-service evaluation | 78.3% | — |
τ²-bench · airlineAgentic tasks — task success in the airline domain of the second-generation customer-service evaluation | 60% | — |
τ²-bench · telecomAgentic tasks — task success in the telecom domain of the second-generation customer-service evaluation | 74.1% | — |
BrowseCompWeb browsing — Can the AI browse the web and track down hard-to-find answers? Higher is better. | — | 63.8% |
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. | — | 47.5% |
AIME 2025Reasoning — single-answer accuracy on the 2025 invitational mathematics exam without tools | 91.1% | — |
IFBenchInstruction following — Tests whether the AI can follow detailed instructions and satisfy multiple constraints at once. Higher is better. | — | 76.1% |
MMLU-ProKnowledge — accuracy on the ten-option MMLU-Pro benchmark | — | 86.7% |
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. | — | 58% |
HealthBench · hardHealthcare — physician rubric score on the hard subset of the original HealthBench evaluation | 40.3% | — |
CharXiv ReasoningChart reasoning — Can the AI read and reason about complex charts and figures, not just text? Higher is better. | — | 77.2% |
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. | 75.5% | — |
MMMU-Pro · standard and vision meanMultimodal reasoning — average across the standard and vision sets | 74.1% | — |
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 (<128K)Long context — breadth-first traversal accuracy for inputs shorter than 128K tokens | 73.4% | — |
GraphWalks parents (<128K)Long context — parent-node retrieval accuracy for inputs shorter than 128K tokens | 64.3% | — |
OpenAI MRCR (2 needles) · 128KLong context — retrieval of two matching turns in the original MRCR evaluation | 84.3% | — |
OpenAI MRCR (2 needles) · 256KLong context — retrieval of two matching turns in the original MRCR evaluation | 58.8% | — |
| Overview | ||
| Company | OpenAI | Qwen |
| Release date | Aug 7 2025 | Feb 24 2026 |
| Access | Closed | Open Weight |
| Model details | View model | View model |
Other comparisons
GPT-5 minivsClaude Fable 5.2Qwen3.5 MediumvsClaude Fable 5.2GPT-5 minivsGemini 4 NeonQwen3.5 MediumvsGemini 4 NeonGPT-5 minivsMuse EmberQwen3.5 MediumvsMuse EmberGPT-5 minivsGrok 4.7Qwen3.5 MediumvsGrok 4.7GPT-5 minivsDeepSeek-V4.1-Flash-1017Qwen3.5 MediumvsDeepSeek-V4.1-Flash-1017GPT-5 minivsMistral Large 4Qwen3.5 MediumvsMistral Large 4Frequently asked questions
Qwen3.5 Medium leads GPT-5 mini on 4 of the 4 benchmarks they both report (SWE-Bench Verified, Humanity's Last Exam, GPQA Diamond, MMMU). Only GPT-5 mini has a verified first-party price for an available API: $0.25 per million input tokens and $2.00 per million output tokens. No currently available pay-as-you-go API rate is tracked for Qwen3.5 Medium. GPT-5 mini shipped 201 days before Qwen3.5 Medium, so benchmark comparisons should account for the intervening progress.
Context windows are 400k (GPT-5 mini) vs 256k (Qwen3.5 Medium). GPT-5 mini is closed, while Qwen3.5 Medium is open weight.
On SWE-Bench Verified, Qwen3.5 Medium leads at 72% vs GPT-5 mini at 71%. On Humanity's Last Exam · no tools, Qwen3.5 Medium leads at 25.3% vs GPT-5 mini at 16.7%. On GPQA Diamond, Qwen3.5 Medium leads at 86.6% vs GPT-5 mini at 82.3%. On MMMU, Qwen3.5 Medium leads at 83.9% vs GPT-5 mini at 81.6%.