Unverified50% confidenceFactExact time
The previous generation v4 Flash was released in July with 284 billion parameters, and the community can pack the 4-bit experts into a 128 GB gguf version that can be run on four second-hand RTX 3060s.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/15/2026
First Seen
Valid until: 12/14/2026
Sources
Related Entities
Related Claims
Partially VerifiedThe 4-bit quantized version (Q4K XL) of Qwen3 Next Flash is approximately 111 GB.76% similarUnverifiedDuring the experiment, DeepSeek V4 Flash released its official version while the Pro version remained in preview, and the Qwen3.8 series had not yet been released, so comparisons used Qwen3.6's 27B and 35B MoE.75% similarUnverifiedA configuration capable of running v4.1 Flash requires 4 NVIDIA DGX Spark units (each equipped with a GB10 Grace Blackwell chip with 128 GB unified memory), priced at $4,699 each, totaling $18,796 for four units.74% similarUnverifiedQwen3-8B Flash performs comparably to DeepSeek's 27B model, has a native context of 256K expandable to 1M, and the official website already provides an API interface with 1 million context.73% similarUnverifiedQwen3.8 Flash Next has a native context window of 262K tokens, which can be extended to 1 million tokens using YaRN technology71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/923767API
curl https://kongchang.com/api/v1/knowledge/claims/923767MCP
get_claim(id=923767)