vrr/kvcache-ai-ktransformers
2
0
Fork 0
mirror of https://github.com/kvcache-ai/ktransformers.git synced 2026-04-30 12:49:52 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 6

Revert "[update] Reorganize documentation/README"

Browse source
This commit is contained in:
Azure 2025-02-15 03:43:48 +08:00 • committed by GitHub
parent 19d4a50b1c
commit 4f4ed36442
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
7 changed files with 241 additions and 420 deletions
Download patch file Download diff file Expand all files Collapse all files

4
doc/en/deepseek-v2-injection.md
View file

@ -1,6 +1,6 @@
# Tutorial: Heterogeneous and Local MoE Inference
# Tutorial: Heterogeneous and Local DeepSeek-V2 Inference
DeepSeek-(Code)-V2 is a series of strong mixture-of-experts (MoE) models, featuring a total of 236 billion parameters, with 21 billion parameters activated per token. This model has demonstrated remarkable reasoning capabilities across various benchmarks, positioning it as one of the SOTA open models and nearly comparable in performance to GPT-4. DeepSeek-R1 uses a similar architecture to DeepSeek-V2, but with a bigger number of parameters.
DeepSeek-(Code)-V2 is a series of strong mixture-of-experts (MoE) models, featuring a total of 236 billion parameters, with 21 billion parameters activated per token. This model has demonstrated remarkable reasoning capabilities across various benchmarks, positioning it as one of the SOTA open models and nearly comparable in performance to GPT-4.
<p align="center">
<picture>