Expired50% confidenceEventExact time
The Miles team, in collaboration with AMD officially, successfully migrated the complete RL training of DeepSeek-V4 Flash to AMD Instinct MI355X GPUs and achieved end-to-end stable operation.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/12/2026
First Seen
This claim has expired. It is kept for the record and no longer describes the current state.
Valid until: 10/10/2026(expired)
Sources
DeepSeek-V4 Flash RL Training Successfully Ported to AMD MI355X GPU: A Complete Breakdown
twitterlmsysorg7/10/2026
Related Claims
UnverifiedIn the CLI40 command-line operation test, DeepSeek V4 Flash performed best, surpassing V4 Pro.66% similarUnverifiedThe DeepSeek V4 Flash free model on OpenRouter performed well in testing, and the automatic failover function was successfully triggered.63% similarUnverifiedAfter repair, DeepSeek V4 Flash can compete directly with flagship models.63% similarUnverifiedA research team successfully completed post-training of the DeepSeek v4 Pro model using Huawei Ascend 910C chips.62% similarUnverifiedAMP（自动混合精度训练）通过将部分计算从 FP32 降至 FP16 可将 GPU 有效计算吞吐量提升 2-3 倍并减少显存占用62% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/490934API
curl https://kongchang.com/api/v1/knowledge/claims/490934MCP
get_claim(id=490934)