Compare · ModelsLive · 2 picked · head to head

Claude Sonnet 4 vs Qwen 3.5 Plus (hosted 397B-A17B)

Side by side · benchmarks, pricing, and signals you can act on.

Winner summary

Qwen 3.5 Plus (hosted 397B-A17B) wins 5 of 5 shared benchmarks. Leads in agentic · math · knowledge.

Category leads
agentic·Qwen 3.5 Plus (hosted 397B-A17B)math·Qwen 3.5 Plus (hosted 397B-A17B)knowledge·Qwen 3.5 Plus (hosted 397B-A17B)
Hype vs Reality
Claude Sonnet 4
#151 by perf·no signal
QUIET
Qwen 3.5 Plus (hosted 397B-A17B)
#202 by perf·no signal
QUIET
Best value
Claude Sonnet 4
4.8 pts/$
$9.00/M
Qwen 3.5 Plus (hosted 397B-A17B)
no price
Vendor risk
Anthropic logo
Anthropic
$380.0B·Tier 1
Medium risk
Alibaba Qwen logo
Alibaba (Qwen)
$293.0B·Tier 1
Low risk
Head to head
Claude Sonnet 4Qwen 3.5 Plus (hosted 397B-A17B)
APEX-Agents
Qwen 3.5 Plus (hosted 397B-A17B) leads by +4.3
APEX-Agents · evaluates AI agents on complex, multi-step tasks requiring planning, tool use, and autonomous decision-making in realistic environments.
Claude Sonnet 4
9.3
Qwen 3.5 Plus (hosted 397B-A17B)
13.6
FrontierMath-2025-02-28-Private
Qwen 3.5 Plus (hosted 397B-A17B) leads by +16.9
FrontierMath (Feb 2025) · original research-level math problems created by mathematicians, testing capabilities at the boundary of current AI mathematical reasoning.
Claude Sonnet 4
4.1
Qwen 3.5 Plus (hosted 397B-A17B)
21.0
FrontierMath-Tier-4-2025-07-01-Private
Qwen 3.5 Plus (hosted 397B-A17B) leads by +2.1
FrontierMath Tier 4 (Jul 2025) · the most challenging tier of frontier mathematics, containing problems that push the absolute limits of AI mathematical reasoning.
Claude Sonnet 4
0.0
Qwen 3.5 Plus (hosted 397B-A17B)
2.1
GPQA diamond
Qwen 3.5 Plus (hosted 397B-A17B) leads by +6.6
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
Claude Sonnet 4
72.3
Qwen 3.5 Plus (hosted 397B-A17B)
78.9
OTIS Mock AIME 2024-2025
Qwen 3.5 Plus (hosted 397B-A17B) leads by +13.9
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
Claude Sonnet 4
71.1
Qwen 3.5 Plus (hosted 397B-A17B)
85.0
Full benchmark table
BenchmarkClaude Sonnet 4Qwen 3.5 Plus (hosted 397B-A17B)
APEX-Agents
APEX-Agents · evaluates AI agents on complex, multi-step tasks requiring planning, tool use, and autonomous decision-making in realistic environments.
9.313.6
FrontierMath-2025-02-28-Private
FrontierMath (Feb 2025) · original research-level math problems created by mathematicians, testing capabilities at the boundary of current AI mathematical reasoning.
4.121.0
FrontierMath-Tier-4-2025-07-01-Private
FrontierMath Tier 4 (Jul 2025) · the most challenging tier of frontier mathematics, containing problems that push the absolute limits of AI mathematical reasoning.
0.02.1
GPQA diamond
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
72.378.9
OTIS Mock AIME 2024-2025
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
71.185.0
Pricing · per 1M tokens · projected $/mo at 10M tokens
ModelInputOutputContextProjected $/mo
Anthropic logoClaude Sonnet 4$3.00$15.001.0M tokens (~500 books)$60.00
Alibaba Qwen logoQwen 3.5 Plus (hosted 397B-A17B)