Presidentlin's picture
x
fffbe5d
import { Benchmark } from "./types";
export const qwenBenchmarks: Benchmark[] = [
{
model: "Qwen3-235B-A22B",
provider: "Qwen",
inputPrice: 0.0,
outputPrice: 0.0,
benchmark: {
aime_24: 85.7,
aime_2025: 81.5,
gpqa_diamond: 44.06,
// livecodebench_v6: 70.7,
mmlu_pro: 68.18,
mmlu: 87.81,
mmmu: 71.84,
// gsm8k: 47.47,
// math: 71.84,
// bigbench_extra_hard: 59.54,
// global_mmlu_lite: 87.40,
// evalplus: 77.60,
// humaneval: 79.00,
// mbpp: 81.40,
// cruxeval_c: 79.00,
simpleqa: 85.8,
egoschema: 81.1,
},
source: "https://qwenlm.github.io/blog/qwen3/",
},
{
model: "Qwen3-32B",
provider: "Qwen",
inputPrice: 0.0,
outputPrice: 0.0,
benchmark: {
aime_24: 81.4,
aime_2025: 72.9,
// livecodebench_v6: 65.7,
//: 1977,
//aider_polyglot: 50.2,
// livebench: 74.9,
// bfcl: 70.3,
// multillm: 73.0,
},
source: "https://qwenlm.github.io/blog/qwe,n3/ (image table)",
},
{
model: "Qwen3-30B-A3B",
provider: "Qwen",
inputPrice: 0.0,
outputPrice: 0.0,
benchmark: {
aime_24: 80.4,
aime_2025: 70.9,
//livecodebench_v6: 62.6,
//codeforces: 1974,
gpqa: 65.8,
// livebench: 74.3,
// bfcl: 69.1,
// multillm: 72.2,
},
source: "https://qwenlm.github.io/blog/qwen3/ (image table)",
},
{
model: "Qwen3-4B",
provider: "Qwen",
inputPrice: 0.0,
outputPrice: 0.0,
benchmark: {
aime_24: 73.8,
aime_2025: 65.6,
//livecodebench_v6: 54.2,
// codeforces: 1671,
gpqa: 55.9,
// bfcl: 65.9,
// math: 62.12,
//multillm: 66.3,
},
source: "https://qwenlm.github.io/blog/qwen3/ (image table)",
}
];