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. | 1M | 1M |
| 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. | $0.10 | — |
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. | $0.40 | — |
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. | $0.01 | — |
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. | $0.05Google AI Studio | — |
Cheapest outputLowest output rate across every provider on OpenRouter, the lab included. May come from a different provider than the cheapest input. | $0.20Google AI Studio | — |
| $40.00Google list price | — | |
| Benchmarks | ||
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. | 33.7% | 38.6% |
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. | 31.6% | — |
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% |
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. | 5.1% | — |
MATHReasoning — accuracy on the original competition mathematics test set | — | 79.5% |
AIME 2025Reasoning — single-answer accuracy on the 2025 invitational mathematics exam without tools | 49.8% | — |
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 64.6% | — |
SimpleQAKnowledge — accuracy on short, fact-seeking questions in the original SimpleQA set | 10.7% | — |
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. | 72.9% | — |
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. | 16.6% | — |
MRCR v2 (8-needle) · 1M pointwiseLong context — Tests whether the AI can find specific details buried inside an enormous document (around 1 million tokens — many books). Higher is better. | 4.1% | — |
| Overview | ||
| Company | Qwen | |
| Release date | Jun 17 2025 | Jan 27 2025 |
| Access | Closed | Open Weight |
| Model details | View model | View model |
Other comparisons
Gemini 2.5 Flash-LitevsClaude Fable 5.2Qwen2.5-1MvsClaude Fable 5.2Gemini 2.5 Flash-LitevsGPT-6.2 AstraQwen2.5-1MvsGPT-6.2 AstraGemini 2.5 Flash-LitevsMuse EmberQwen2.5-1MvsMuse EmberGemini 2.5 Flash-LitevsGrok 4.7Qwen2.5-1MvsGrok 4.7Gemini 2.5 Flash-LitevsDeepSeek-V4.1-Flash-1017Qwen2.5-1MvsDeepSeek-V4.1-Flash-1017Gemini 2.5 Flash-LitevsMistral Large 4Qwen2.5-1MvsMistral Large 4Frequently asked questions
Qwen2.5-1M leads Gemini 2.5 Flash-Lite on 1 of the 1 benchmark they both report (LiveCodeBench). Only Gemini 2.5 Flash-Lite has a verified first-party price for an available API: $0.10 per million input tokens and $0.40 per million output tokens. No currently available pay-as-you-go API rate is tracked for Qwen2.5-1M. Qwen2.5-1M shipped 141 days before Gemini 2.5 Flash-Lite, so benchmark comparisons should account for the intervening progress.
Context windows are 1M (Gemini 2.5 Flash-Lite) vs 1M (Qwen2.5-1M). Gemini 2.5 Flash-Lite is closed, while Qwen2.5-1M is open weight.
On LiveCodeBench, Qwen2.5-1M leads at 38.6% vs Gemini 2.5 Flash-Lite at 33.7%.