Qwen: Qwen3 30B A3B Instruct 2507
Released Jul 29, 2025Ranked in:Text
Try it on:
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...
Best overall rank
#141
▼ 1 vs 7d ago
Total arena votes
127,868
Across 27 (subset, category) entries
Providers on OR
6
WandB + others
First seen
2026-05-08
45 days tracked
Trend
Pricing
Per token
- Input (per 1M)
- $0.048
- Output (per 1M)
- $0.193
Capabilities
Modality
text->textContext
131K tokens
Tokenizer
Qwen3
Knowledge cutoff
2025-06-30
Max output
32K tokens
Moderation
Unmoderated
Hugging Face
Supported parameters
frequency_penalty
logit_bias
logprobs
max_tokens
min_p
presence_penalty
repetition_penalty
response_format
seed
stop
structured_outputs
temperature
tool_choice
tools
top_k
top_logprobs
top_p
Providers
| Provider | Region / Tag | Context | Input/M | Output/M | Uptime 1d | Latency | Quant | Cache |
|---|---|---|---|---|---|---|---|---|
| WandB | wandb/bf16 | 262K | $0.100 | $0.300 | 100.00% | - | bf16 | - |
| AtlasCloud | atlas-cloud/fp8 | 131K | $0.100 | $0.300 | 99.93% | - | fp8 | - |
| StreamLake | streamlake | 128K | $0.048 | $0.193 | 99.84% | - | - | - |
| Nebius | nebius/fp8 | 262K | $0.100 | $0.300 | 99.58% | - | fp8 | - |
| Alibaba | alibaba | 131K | $0.130 | $0.520 | 99.33% | - | - | - |
| SiliconFlow | siliconflow/fp8 | 262K | $0.090 | $0.300 | 84.08% | - | fp8 | - |
Per-category ranks
Text
| Rank | Category | Variant | Trend | Elo | Votes |
|---|---|---|---|---|---|
| #109 | Chinese | - | 1438 | 1,326 | |
| #109 | French | - | 1421 | 306 | |
| #110 | Japanese | - | 1325 | 704 | |
| #117 | Spanish | - | 1386 | 632 | |
| #123 | Coding | - | 1440 | 4,658 | |
| #123 | Industry Business And Management And Financial Operations | - | 1397 | 4,175 | |
| #124 | Industry Software And It Services | - | 1430 | 8,196 | |
| #126 | Industry Mathematical | - | 1401 | 1,205 | |
| #133 | German | - | 1359 | 487 | |
| #136 | Hard Prompts | - | 1407 | 11,068 | |
| #138 | Hard Prompts English | - | 1416 | 5,623 | |
| #138 | Korean | - | 1314 | 566 | |
| #140 | Exclude Ties | - | 1359 | 16,690 | |
| #141 | Overall | - | 1383 | 23,728 | |
| #142 | Multi Turn | - | 1383 | 4,072 | |
| #143 | Expert | - | 1393 | 1,144 | |
| #143 | Industry Medicine And Healthcare | - | 1399 | 1,346 | |
| #144 | English | - | 1393 | 11,050 | |
| #144 | Longer Query | - | 1382 | 4,880 | |
| #145 | Industry Life And Physical And Social Science | - | 1394 | 3,679 | |
| #145 | Math | - | 1381 | 1,426 | |
| #146 | Instruction Following | - | 1367 | 5,963 | |
| #146 | Russian | - | 1366 | 1,227 | |
| #148 | Industry Legal And Government | - | 1382 | 1,491 | |
| #153 | Industry Entertainment And Sports And Media | - | 1333 | 4,144 | |
| #160 | Industry Writing And Literature And Language | - | 1343 | 5,098 | |
| #170 | Creative Writing | - | 1321 | 2,984 |