Expired50% confidenceFactExact time
After completing both the Super Mario game and backend framework page tasks, MiniMax M3's token consumption was lower than DeepSeek's.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/2/2026
First Seen
Valid until: 9/30/2026(expired)
Sources
MiniMax M3 Code Test: How Does It Stack Up Against DeepSeek?
bilibili程序员老罗6/5/2026
Related Claims
UnverifiedMiniMax has advantages in long-context processing.63% similarUnverifiedMiniMax M1 can process a context window of up to 1 million tokens in a single pass.56% similarUnverifiedDeepSeek V4 uses compression techniques such as MLA to reduce KV Cache VRAM usage to below 10% of the previous generation.55% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/46901API
curl https://kongchang.com/api/v1/knowledge/claims/46901MCP
get_claim(id=46901)