Unverified50% confidenceFactExact time
MNN-Pocket-0.3B supports a context length of up to 32K tokens.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
11/10/2026
First Seen
Valid until: 2/8/2027
Sources
Related Entities
Related Claims
UnverifiedMNN-Pocket-0.3B is a 300M-parameter dense model.70% similarUnverifiedUnder high power operation, the 35B MoE model reaches 78 to 107 tokens/s, while the dense 27B model only reaches 22 tokens/s.67% similarVerifiedTaking LLaMA-2 70B as an example, when processing a 4096-token context with a batch size of 32, the KV Cache can consume over 80 GB of memory.67% similarUnverifiedThe baseline BF16 model reaches about 37.5 tokens/s at single concurrency, 714 tokens/s aggregate throughput at 32 concurrent requests, and 61ms first-token latency; after switching to FP8 weights, single concurrency rises to 53.4 tokens/s, and peak throughput at 64 concurrency rises from about 800 to 1,140 tokens/s, about 1.42x.67% similarUnverifiedWhen running Qwen3-Next-Flash (Autoround W4A16 quantization), the prefill speed was approximately 1.3k tokens/s, and the generation speed reached 70 tokens/s (code) / 60 tokens/s (prose), with support for 128k+ long context.67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1154761API
curl https://kongchang.com/api/v1/knowledge/claims/1154761MCP
get_claim(id=1154761)