DeepSeek-R1-0528
Listed in tt-inference-server on 3 hardware types; best status Experimental.
In tt-inference-server
| Hardware | Engine and implementation | Status | Context | Users | Build | Support page |
|---|---|---|---|---|---|---|
Dual WH GalaxyDUAL_GALAXY |
vLLM deepseek-r1-galaxy |
Experimental | 32K | 256 | tt-metal 805f43dvLLM a45c614Image ghcr.io/tenstorrent/tt-inference-server/vllm-tt-metal-src-multihost-ubuntu-22.04-amd64:0.12.0-805f43d-a45c614 |
Open |
Quad WH GalaxyQUAD_GALAXY |
vLLM deepseek-r1-galaxy |
Experimental | 32K | 512 | tt-metal 805f43dvLLM a45c614Image ghcr.io/tenstorrent/tt-inference-server/vllm-tt-metal-src-multihost-ubuntu-22.04-amd64:0.12.0-805f43d-a45c614 |
Open |
WH GalaxyGALAXY |
vLLM deepseek-r1-galaxy |
Experimental | 2K | 256 | tt-metal 805f43dvLLM a45c614Image ghcr.io/tenstorrent/tt-inference-server/vllm-tt-metal-src-release-ubuntu-22.04-amd64:0.12.0-805f43d-a45c614 |
Open |
Status, context and users are copied from tt-inference-server's release spec (commit 2db98e38c, release 0.25.0). Context is the longest prompt plus reply the entry allows. Users is its highest concurrency. Each row carries its own code link, tt-metal and vLLM commits and image, because they differ between hardware. Nothing on this page is measured.
tt-model bundles
No tt-model bundle has been published for this model.
Details
| Kind | LLM (from tt-inference-server model type) |
| Family | DeepSeek (from the start of the name) |
| Implementation | deepseek-r1-galaxy |
| Tool-call parser | deepseek_v3 |
| Reasoning parser | deepseek_r1 |
| Listed by | tt-inference-server |