Expired50% confidenceBenchmarkExact time
At the time of its release, DeepSeek-V3 matched or surpassed Claude 3.5 Sonnet on multiple benchmarks, with training costs reportedly a fraction of those of mainstream closed-source models.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/7/2026
First Seen
Valid until: 10/5/2026(expired)
Sources
How Anthropic Gradually Lost Developer Trust: Pricing, Transparency, and Vendor Lock-in
hackernewshackernews7/6/2026
Related Claims
UnverifiedBased on comparative testing, DeepSeek V4 Pro's programming performance even surpasses Claude Sonnet, and its price is much cheaper.75% similarUnverifiedGLM-4.7 surpasses DeepSeek V3.2 and Claude Sonnet 4.5 in code capability benchmarks.71% similarUnverifiedThe invocation cost of some open-source or lightweight models (such as Mistral and the Llama series via third-party hosting) may be as low as one-tenth of Claude 3.5 Sonnet's cost or even lower.70% similarUnverifiedThe gap between DeepSeek and Claude Sonnet on the SWE-bench coding benchmark is within 1%67% similarVerifiedClaude 3.5 Sonnet has long led in code benchmarks such as SWE-bench.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/170418API
curl https://kongchang.com/api/v1/knowledge/claims/170418MCP
get_claim(id=170418)