Unverified50% confidenceBenchmarkExact time
In the InferenceX public benchmark, Jalapeño achieved 1.5 to 1.9 times the tasks per watt at peak throughput compared to NVIDIA's reference system, with end-to-end latency reduced by 1.7 to 3.6 times.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/24/2026
First Seen
Valid until: 12/23/2026
Sources
Related Entities
Related Claims
UnverifiedBullet 在 SWE-bench Verified 基准测试中取得 95.8% 的成绩，位列前三，平均每个任务耗时 119 秒61% similarUnverifiedMuse Glimmer 30B is reported to achieve 74.9 tokens per second on an RTX 5090, and reviewers observed approximately 60–65 tokens/sec running the full-precision version on an RTX 3090.60% similarUnverifiedIn the GPQA-Diamond xhigh test, the base model Qwen3.8-27B achieves 88.4% accuracy consuming 6,642 median tokens, while Swift maintains 88.3% accuracy with tokens reduced to 2,77159% similarUnverifiedAfter full pretraining, NCP-ArchPreview outperforms OLMo-3-7B by 2.45 points on the macro-average score across downstream tasks.59% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/945919API
curl https://kongchang.com/api/v1/knowledge/claims/945919MCP
get_claim(id=945919)