Qwen: Qwen3 30B A3B Instruct 2507

Description

Qwen3-30B-A3B-Instruct-2507 is a 30.5B-parameter mixture-of-experts language model from Qwen, with 3.3B active parameters per inference. It operates in non-thinking mode and is designed for high-quality instruction following, multilingual understanding, and agentic tool use. Post-trained on instruction data, it demonstrates competitive performance across reasoning (AIME, ZebraLogic), coding (MultiPL-E, LiveCodeBench), and alignment (IFEval, WritingBench) benchmarks. It outperforms its non-instruct variant on subjective and open-ended tasks while retaining strong factual and coding performance.

Raw fields snapshot

{
  "id": "qwen/qwen3-30b-a3b-instruct-2507",
  "name": "Qwen: Qwen3 30B A3B Instruct 2507",
  "description": "Qwen3-30B-A3B-Instruct-2507 is a 30.5B-parameter mixture-of-experts language model from Qwen, with 3.3B active parameters per inference. It operates in non-thinking mode and is designed for high-quality instruction following, multilingual understanding, and agentic tool use. Post-trained on instruction data, it demonstrates competitive performance across reasoning (AIME, ZebraLogic), coding (MultiPL-E, LiveCodeBench), and alignment (IFEval, WritingBench) benchmarks. It outperforms its non-instruct variant on subjective and open-ended tasks while retaining strong factual and coding performance.",
  "created": 1753806965,
  "canonical_slug": "qwen/qwen3-30b-a3b-instruct-2507",
  "hugging_face_id": "Qwen/Qwen3-30B-A3B-Instruct-2507",
  "source_type": "both",
  "context_length": 262144,
  "max_completion_tokens": 262144,
  "is_moderated": false,
  "architecture": {
    "modality": "text->text",
    "input_modalities": [
      "text"
    ],
    "output_modalities": [
      "text"
    ],
    "tokenizer": "Qwen3",
    "instruct_type": null
  },
  "input_modalities": [
    "text"
  ],
  "output_modalities": [
    "text"
  ],
  "modality": "text->text",
  "tokenizer": "Qwen3",
  "instruct_type": null,
  "supported_parameters": [
    "frequency_penalty",
    "max_tokens",
    "presence_penalty",
    "repetition_penalty",
    "response_format",
    "seed",
    "structured_outputs",
    "temperature",
    "tool_choice",
    "tools",
    "top_k",
    "top_p"
  ],
  "default_parameters": {},
  "per_request_limits": null,
  "top_provider": {
    "context_length": 262144,
    "max_completion_tokens": 262144,
    "is_moderated": false
  },
  "pricing": {
    "prompt": "0.00000009",
    "completion": "0.0000003"
  },
  "PPM": {
    "prompt": 0.09,
    "completion": 0.3
  },
  "openrouter_raw": {
    "id": "qwen/qwen3-30b-a3b-instruct-2507",
    "canonical_slug": "qwen/qwen3-30b-a3b-instruct-2507",
    "hugging_face_id": "Qwen/Qwen3-30B-A3B-Instruct-2507",
    "name": "Qwen: Qwen3 30B A3B Instruct 2507",
    "created": 1753806965,
    "description": "Qwen3-30B-A3B-Instruct-2507 is a 30.5B-parameter mixture-of-experts language model from Qwen, with 3.3B active parameters per inference. It operates in non-thinking mode and is designed for high-quality instruction following, multilingual understanding, and agentic tool use. Post-trained on instruction data, it demonstrates competitive performance across reasoning (AIME, ZebraLogic), coding (MultiPL-E, LiveCodeBench), and alignment (IFEval, WritingBench) benchmarks. It outperforms its non-instruct variant on subjective and open-ended tasks while retaining strong factual and coding performance.",
    "context_length": 262144,
    "architecture": {
      "modality": "text->text",
      "input_modalities": [
        "text"
      ],
      "output_modalities": [
        "text"
      ],
      "tokenizer": "Qwen3",
      "instruct_type": null
    },
    "pricing": {
      "prompt": "0.00000009",
      "completion": "0.0000003"
    },
    "top_provider": {
      "context_length": 262144,
      "max_completion_tokens": 262144,
      "is_moderated": false
    },
    "per_request_limits": null,
    "supported_parameters": [
      "frequency_penalty",
      "max_tokens",
      "presence_penalty",
      "repetition_penalty",
      "response_format",
      "seed",
      "structured_outputs",
      "temperature",
      "tool_choice",
      "tools",
      "top_k",
      "top_p"
    ],
    "default_parameters": {},
    "expiration_date": null
  },
  "categories": {
    "text-expert": {
      "score": 1391,
      "rank_ub": 93,
      "votes": 1150,
      "ci_95": "±17"
    },
    "text-industry-software-and-it-services": {
      "score": 1429,
      "rank_ub": 77,
      "votes": 8224,
      "ci_95": "±7"
    },
    "text-industry-writing-and-literature-and-language": {
      "score": 1343,
      "rank_ub": 111,
      "votes": 5181,
      "ci_95": "±8"
    },
    "text-overall": {
      "score": 1384,
      "rank_ub": 94,
      "votes": 23940,
      "ci_95": "±5"
    },
    "text-industry-life-and-physical-and-social-science": {
      "score": 1395,
      "rank_ub": 98,
      "votes": 3722,
      "ci_95": "±10"
    },
    "text-industry-mathematical": {
      "score": 1406,
      "rank_ub": 78,
      "votes": 1218,
      "ci_95": "±17"
    },
    "text-industry-entertainment-and-sports-and-media": {
      "score": 1334,
      "rank_ub": 106,
      "votes": 4202,
      "ci_95": "±9"
    },
    "text-industry-business-and-management-and-financial-operations": {
      "score": 1396,
      "rank_ub": 74,
      "votes": 4172,
      "ci_95": "±9"
    },
    "text-industry-medicine-and-healthcare": {
      "score": 1400,
      "rank_ub": 96,
      "votes": 1360,
      "ci_95": "±16"
    },
    "text-industry-legal-and-government": {
      "score": 1381,
      "rank_ub": 104,
      "votes": 1509,
      "ci_95": "±15"
    },
    "text-instruction-following": {
      "score": 1367,
      "rank_ub": 99,
      "votes": 5988,
      "ci_95": "±8"
    },
    "text-math": {
      "score": 1386,
      "rank_ub": 97,
      "votes": 1433,
      "ci_95": "±15"
    },
    "text-creative-writing": {
      "score": 1325,
      "rank_ub": 118,
      "votes": 3039,
      "ci_95": "±11"
    },
    "text-multi-turn": {
      "score": 1384,
      "rank_ub": 94,
      "votes": 4082,
      "ci_95": "±9"
    },
    "text-hard-prompts": {
      "score": 1408,
      "rank_ub": 86,
      "votes": 11112,
      "ci_95": "±6"
    },
    "text-coding": {
      "score": 1440,
      "rank_ub": 77,
      "votes": 4661,
      "ci_95": "±9"
    },
    "text-hard-prompts-english": {
      "score": 1417,
      "rank_ub": 88,
      "votes": 5652,
      "ci_95": "±8"
    },
    "text-longer-query": {
      "score": 1382,
      "rank_ub": 97,
      "votes": 4918,
      "ci_95": "±9"
    },
    "text-english": {
      "score": 1395,
      "rank_ub": 96,
      "votes": 11170,
      "ci_95": "±6"
    },
    "text-chinese": {
      "score": 1438,
      "rank_ub": 61,
      "votes": 1323,
      "ci_95": "±17"
    },
    "text-french": {
      "score": 1411,
      "rank_ub": 65,
      "votes": 317,
      "ci_95": "±33"
    },
    "text-spanish": {
      "score": 1387,
      "rank_ub": 70,
      "votes": 641,
      "ci_95": "±25"
    },
    "text-russian": {
      "score": 1366,
      "rank_ub": 96,
      "votes": 1221,
      "ci_95": "±16"
    },
    "text-german": {
      "score": 1351,
      "rank_ub": 86,
      "votes": 487,
      "ci_95": "±27"
    },
    "text-japanese": {
      "score": 1322,
      "rank_ub": 67,
      "votes": 714,
      "ci_95": "±24"
    },
    "text-korean": {
      "score": 1314,
      "rank_ub": 87,
      "votes": 548,
      "ci_95": "±27"
    },
    "text-exclude-ties": {
      "score": 1360,
      "rank_ub": 92,
      "votes": 16808,
      "ci_95": "±7"
    }
  },
  "arena_model_id": "qwen3-30b-a3b-instruct-2507",
  "leaderboard_name": "qwen3-30b-a3b-instruct-2507",
  "match_method": "openrouter_name",
  "match_key": "qwen3-30b-a3b-instruct-2507",
  "match_input": "Qwen: Qwen3 30B A3B Instruct 2507",
  "arena_aliases": []
}