Model AI Qwen3 Max (Preview) dari Alibaba.
Quality
19.2
Coding
65.1
Input Cost
$1.20
Speed
61 tok/s
Perbandingan cepat
Grafik perbandingan kemampuan
Setiap tab membandingkan Qwen3 Max (Preview) dengan 10 model terbaik untuk metrik tersebut, bukan daftar model yang sama.
Quality
Artificial Analysis Intelligence Index
19.2 Qwen3 Max (Preview) 100 Apertus 8B Instruct 100 Gemma 3 1B Instruct 100 Gemma 3n E2B Instruct 100 Granite 4.0 350M 100 Granite 4.0 H 350M 100 LFM2.5-VL-1.6B 100 Qwen3 0.6B (Non-reas… 100 Tiny Aya Global 61 Llama 4 70B 59.9 Claude Fable 5 (Adap…
Qwen3 Max (Preview)
Saat ini Apertus 8B Instruct
Gemma 3 1B Instruct
Gemma 3n E2B Instruct
Granite 4.0 350M
Granite 4.0 H 350M
LFM2.5-VL-1.6B
Qwen3 0.6B (Non-reas…
Tiny Aya Global
Llama 4 70B
Claude Fable 5 (Adap…
Reasoning
GPQA / HLE proxy
76.4 Qwen3 Max (Preview) 94.1 Gemini 3.1 Pro Preview 93.5 GPT-5.5 (xhigh) 93.2 GPT-5.5 (high) 92.9 MiniMax-M3 92.6 Claude Fable 5 (Adap… 92.6 GPT-5.5 (medium) 92.3 Qwen3.7 Max 92.2 Gemini 3.5 Flash (hi… 92.1 Gemini 3.5 Flash (me… 92 Claude Opus 4.8 (Ada…
Qwen3 Max (Preview)
Saat ini Gemini 3.1 Pro Preview
GPT-5.5 (xhigh)
GPT-5.5 (high)
MiniMax-M3
Claude Fable 5 (Adap…
GPT-5.5 (medium)
Qwen3.7 Max
Gemini 3.5 Flash (hi…
Gemini 3.5 Flash (me…
Claude Opus 4.8 (Ada…
Coding
Coding index / coding benchmarks
65.1 Qwen3 Max (Preview) 91.7 Gemini 3 Pro Preview… 90.8 Gemini 3 Flash Previ… 89.6 DeepSeek V3.2 Speciale 89.4 GLM-4.7 (Reasoning) 89.4 GPT-5.2 (medium) 88.9 GPT-5.2 (xhigh) 87.1 Claude Opus 4.5 (Rea… 86.8 GPT-5.1 (high) 86.8 MiMo-V2-Flash (Reaso… 86.2 DeepSeek V3.2 (Reaso…
Qwen3 Max (Preview)
Saat ini Gemini 3 Pro Preview…
Gemini 3 Flash Previ…
DeepSeek V3.2 Speciale
GLM-4.7 (Reasoning)
GPT-5.2 (medium)
GPT-5.2 (xhigh)
Claude Opus 4.5 (Rea…
GPT-5.1 (high)
MiMo-V2-Flash (Reaso…
DeepSeek V3.2 (Reaso…
Math
Math benchmark score
75 Qwen3 Max (Preview) 99 GPT-5.2 (xhigh) 98.7 GPT-5 Codex (high) 98.6 Gemini 2.5 Pro Previ… 98.5 o3-mini (high) 98.1 Gemini 2.5 Flash Pre… 98 Gemini 2.5 Pro Previ… 97.3 o3-mini 97 Gemini 3 Flash Previ… 97 o1 96.7 DeepSeek V3.2 Speciale
Qwen3 Max (Preview)
Saat ini GPT-5.2 (xhigh)
GPT-5 Codex (high)
Gemini 2.5 Pro Previ…
o3-mini (high)
Gemini 2.5 Flash Pre…
Gemini 2.5 Pro Previ…
o3-mini
Gemini 3 Flash Previ…
o1
DeepSeek V3.2 Speciale
Input Cost
USD per 1M input token
$1.20 Qwen3 Max (Preview) $0.00 Apriel-v1.5-15B-Thin… $0.00 Apriel-v1.6-15B-Thin… $0.00 Arctic Instruct $0.00 Claude 2.0 $0.00 Claude 2.1 $0.00 Claude 3.7 Sonnet (R… $0.00 Claude Instant $0.00 Command A+ $0.00 DBRX Instruct $0.00 DeepHermes 3 - Llama…
Qwen3 Max (Preview)
Saat ini Apriel-v1.5-15B-Thin…
Apriel-v1.6-15B-Thin…
Arctic Instruct
Claude 2.0
Claude 2.1
Claude 3.7 Sonnet (R…
Claude Instant
Command A+
DBRX Instruct
DeepHermes 3 - Llama…
Output Cost
USD per 1M output token
$6.00 Qwen3 Max (Preview) $0.00 Apriel-v1.5-15B-Thin… $0.00 Apriel-v1.6-15B-Thin… $0.00 Arctic Instruct $0.00 Claude 2.0 $0.00 Claude 2.1 $0.00 Claude 3.7 Sonnet (R… $0.00 Claude Instant $0.00 Command A+ $0.00 DBRX Instruct $0.00 DeepHermes 3 - Llama…
Qwen3 Max (Preview)
Saat ini Apriel-v1.5-15B-Thin…
Apriel-v1.6-15B-Thin…
Arctic Instruct
Claude 2.0
Claude 2.1
Claude 3.7 Sonnet (R…
Claude Instant
Command A+
DBRX Instruct
DeepHermes 3 - Llama…
Speed
Median output token/sec
61 tok/s Qwen3 Max (Preview) 1002 tok/s Mercury 2 566 tok/s LFM2 1.2B 549 tok/s LFM2.5-1.2B-Instruct 477 tok/s Granite 4.0 H Small 442 tok/s LFM2.5-VL-1.6B 395 tok/s Step 3.7 Flash 376 tok/s HyperNova 60B 2605 368 tok/s Granite 3.3 8B (Non-… 359 tok/s gpt-oss-120b (low) 353 tok/s Gemini 2.5 Flash-Lit…
Qwen3 Max (Preview)
Saat ini Mercury 2
LFM2 1.2B
LFM2.5-1.2B-Instruct
Granite 4.0 H Small
LFM2.5-VL-1.6B
Step 3.7 Flash
HyperNova 60B 2605
Granite 3.3 8B (Non-…
gpt-oss-120b (low)
Gemini 2.5 Flash-Lit…
TTFT
Median time to first token
1.9s Qwen3 Max (Preview) 0ms Apertus 70B Instruct 0ms Apertus 8B Instruct 0ms Apriel-v1.5-15B-Thin… 0ms Apriel-v1.6-15B-Thin… 0ms Arctic Instruct 0ms Claude 2.0 0ms Claude 2.1 0ms Claude 3.5 Haiku 0ms Claude 3.7 Sonnet (N… 0ms Claude 3.7 Sonnet (R…
Qwen3 Max (Preview)
Saat ini Apertus 70B Instruct
Apertus 8B Instruct
Apriel-v1.5-15B-Thin…
Apriel-v1.6-15B-Thin…
Arctic Instruct
Claude 2.0
Claude 2.1
Claude 3.5 Haiku
Claude 3.7 Sonnet (N…
Claude 3.7 Sonnet (R…
Spesifikasi Qwen3 Max (Preview)
- Quality
- 19.2
- Reasoning
- 76.4
- Coding
- 65.1
- Math
- 75
- Knowledge
- 83.8
- Instruction Following
- 48
- Tool Use
- 32.7
- Input Cost
- $1.20
- Output Cost
- $6.00
- Blended Cost
- $2.40
- Speed
- 61 tok/s
- TTFT
- 1.9s
- Provider
- Alibaba
- Release Date
- -
Review & Deskripsi Lengkap
Rekomendasi lanjut
Model AI terkait
Swiss AI Initiative
Apertus 8B Instruct
Apertus 8B Instruct adalah model LLM dari Swiss AI Initiative. Data tersedia: skor kualitas, reasoning, coding, matematika, biaya token, speed, TTFT, dan tanggal rilis dari Artificial Analysis.
GoogleGemma 3 1B Instruct
Gemma 3 1B Instruct adalah model LLM dari Google. Data tersedia: skor kualitas, reasoning, coding, matematika, biaya token, speed, TTFT, dan tanggal rilis dari Artificial Analysis.
GoogleGemma 3n E2B Instruct
Gemma 3n E2B Instruct adalah model LLM dari Google. Data tersedia: skor kualitas, reasoning, coding, matematika, biaya token, speed, TTFT, dan tanggal rilis dari Artificial Analysis.
IBMGranite 4.0 350M
Granite 4.0 350M adalah model LLM dari IBM. Data tersedia: skor kualitas, reasoning, coding, matematika, biaya token, speed, TTFT, dan tanggal rilis dari Artificial Analysis.
IBMGranite 4.0 H 350M
Granite 4.0 H 350M adalah model LLM dari IBM. Data tersedia: skor kualitas, reasoning, coding, matematika, biaya token, speed, TTFT, dan tanggal rilis dari Artificial Analysis.
Liquid AILFM2.5-VL-1.6B
LFM2.5-VL-1.6B adalah model LLM dari Liquid AI. Data tersedia: skor kualitas, reasoning, coding, matematika, biaya token, speed, TTFT, dan tanggal rilis dari Artificial Analysis.