Filter by type:
Filter by tags:
dfm-mimir:vllm
DFM Mimir is an Apache-2.0, instruction-tuned HRM-Text model from Danish
Foundation Models. It has about 1 billion parameters and a 4,096-token
context window. The model focuses on Danish and English chat, reasoning,
mathematics, and code generation, and uses only permissible post-training
data. This entry serves the official BF16 safetensors checkpoint with vLLM.
Repository: localaiLicense: apache-2.0
vllm-omni-z-image-turbo
Z-Image-Turbo via vLLM-Omni - A distilled version of Z-Image optimized for speed with only 8 NFEs. Offers sub-second inference latency on enterprise-grade H800 GPUs and fits within 16GB VRAM. Excels in photorealistic image generation, bilingual text rendering (English & Chinese), and robust instruction adherence.
Repository: localaiLicense: apache-2.0
vllm-omni-wan2.2-t2v
Wan2.2-T2V-A14B via vLLM-Omni - Text-to-video generation model from Wan-AI. Generates high-quality videos from text prompts using a 14B parameter diffusion model.
Repository: localaiLicense: apache-2.0
vllm-omni-wan2.2-i2v
Wan2.2-I2V-A14B via vLLM-Omni - Image-to-video generation model from Wan-AI. Generates high-quality videos from images using a 14B parameter diffusion model.
Repository: localaiLicense: apache-2.0
qwen3.8-27b-exl3-vllm-cpp
Qwen3.8-27B EXL3 3.5bpw served by vllm.cpp, LocalAI's C++ vLLM-style
runtime. The published checkpoint generates on CUDA and measured 16.7
tokens/s on GB10 with the pinned revision and the limits configured here.
This is the target-only setup. Use the DFlash2 variant for the measured
speculative-decoding configuration. The entry downloads the complete
revision-pinned repository, including its configuration, tokenizer, index,
and safetensors shards.
Repository: localai
qwen3.8-27b-dflash2-exl3-vllm-cpp
Qwen3.8-27B EXL3 with its EXL3 DFlash2 companion, served by vllm.cpp. On
GB10 this pinned pair measured 48.7 tokens/s at a seven-token draft budget,
versus 16.7 tokens/s target-only, with token-identical greedy output.
LocalAI stages both complete Hugging Face repositories before load. The
content-addressed companion snapshot is passed to the backend as the draft
model, while the speculative method and seven-token budget remain fixed.
Repository: localai
deepseek-v4-flash-spark-exl3-vllm-cpp
DeepSeek V4 Flash's Spark and GB10-oriented REAP-K216 EXL3 checkpoint,
served by vllm.cpp. It needs CUDA and roughly 100 GiB for its large
rank-sliced checkpoint.
The repository is pinned to its latest recorded revision. vllm.cpp's
existing runtime evidence measured the older 22f28d32b9b29b4352eaa380ff8c2c170b2847ab
revision; this entry does not claim that the newer revision has passed the
same end-to-end gate.
Repository: localai
deepseek-v4-flash-exl3-3bpw-vllm-cpp
Experimental non-Spark DeepSeek V4 Flash EXL3 3.0bpw checkpoint served by
vllm.cpp. This is the complete, non-REAP layout and requires a large
multi-GPU CUDA system.
The publisher describes the artifact as structurally complete but has not
passed end-to-end generation. Treat this entry as an integration target,
not as a correctness- or performance-gated configuration.
Repository: localai
qwen3.6-27b-nvfp4-vllm-cpp
Qwen3.6-27B in NVFP4, served by vllm.cpp: LocalAI's own C++ port of vLLM, with
no Python at inference time. This is the reference text-generation checkpoint
the engine is gated on, token-for-token identical to vLLM's own greedy output
over the 235-prompt correctness battery, and measured at or above vLLM's
throughput at every concurrency from 1 to 32.
The weights are PINNED to revision 890bdef7. That pin is load-bearing, not
housekeeping: the same repository name was later re-quantized to FP8 W8A8
throughout, so an unpinned copy of this entry serves entirely different
weights with no error and none of the measured behaviour above.
Needs a Blackwell-class NVIDIA GPU (NVFP4 has no kernel on older
architectures) and roughly 25 GB of weights plus KV cache. Tool calling and
the thinking split are parsed inside the engine.
Repository: localaiLicense: apache-2.0
qwen3.6-27b-nvfp4-mtp-vllm-cpp
Qwen3.6-27B NVFP4 on vllm.cpp with MTP speculative decoding enabled.
MTP (Multi-Token Prediction) drafts from a head that ships inside the target
checkpoint's own mtp.* tensors, so there is no second model to download and
no extra weights to manage. The verifier accepts roughly 85% of drafted
tokens on prose and 92% on code, worth about 1.5x to 1.6x the decode
throughput of the same weights with speculation off, and it holds that lead
at concurrency 2, 4 and 8.
Same weights and same revision pin as qwen3.6-27b-nvfp4-vllm-cpp; install
that entry instead if you would rather not spend the extra memory. The
speculative state (a doubled recurrent-state slot plus the draft cache and
head) costs roughly 3.6 GB on top of the base footprint.
MTP here is depth 1 by construction: the engine refuses
num_speculative_tokens above 1 for this method.
Repository: localaiLicense: apache-2.0
qwen3.6-27b-nvfp4-dflash-vllm-cpp
Qwen3.6-27B NVFP4 on vllm.cpp with DFlash block-diffusion speculative
decoding: the fastest configuration of this model the engine ships.
Where MTP drafts one token at a time, DFlash drafts a whole 16-token block in
a single non-autoregressive pass from a separate 3.5 GB drafter, then the
target verifies the block in one step. At concurrency 1 that measures 2.9x
the throughput of the same weights with speculation off, and at or above
vLLM's own DFlash-on decode.
Both checkpoints are installed for you: the target as a revision-pinned
snapshot, the drafter into models/Qwen3.6-27B-DFlash, which is where the
backend looks when speculative_config.model names it. The drafter shares the
target's embed_tokens and lm_head, so the two are not independently
swappable.
Needs a Blackwell-class NVIDIA GPU and roughly 28 GB of weights in total.
Repository: localaiLicense: apache-2.0
qwen3.6-35b-a3b-nvfp4-vllm-cpp
Qwen3.6-35B-A3B in NVFP4, served by vllm.cpp. A 35B mixture-of-experts model
with roughly 3B parameters active per token, so it reads like a much larger
model while costing about as much per token as a small one.
Image input is implemented in the engine but is not token-gated against
vLLM yet, so the vision usecase on this entry is experimental.
This is the engine's gated MoE checkpoint: token-for-token identical to vLLM
over the 315-prompt battery on both the synchronous and asynchronous paths,
at 0.92x to 0.97x vLLM's throughput from concurrency 1 to 32.
The architecture is a gated-delta-net hybrid, so automatic prefix caching is
off by default here where it would be on for a dense model. That is the
engine's own default and this entry does not override it.
Needs a Blackwell-class NVIDIA GPU and roughly 23 GB of weights plus KV cache.
Repository: localaiLicense: apache-2.0
qwen3.6-35b-a3b-nvfp4-mtp-vllm-cpp
Qwen3.6-35B-A3B NVFP4 on vllm.cpp with MTP speculative decoding enabled.
Image input is implemented in the engine but is not token-gated against
vLLM yet, so the vision usecase on this entry is experimental.
The draft head ships inside the checkpoint's own mtp.* tensors, so there is
no second model to download. On this model the speculative path is
token-exact against speculation-off on both the synchronous and asynchronous
schedulers.
Same weights as qwen3.6-35b-a3b-nvfp4-vllm-cpp; install that entry instead if
you would rather not spend the extra memory on speculative state. MTP is
depth 1 by construction on this engine.
Repository: localaiLicense: apache-2.0
qwen3-coder-30b-a3b-vllm-cpp
Qwen3-Coder-30B-A3B on vllm.cpp: a coding and agentic-tool-use model, 30B
total parameters with about 3B active per token, gated token-exact against
vLLM on this engine.
The tool-call parser is named explicitly rather than auto-detected, and that
matters here. Qwen3-Coder's tool dialect is byte-identical on the wire to
another family's, so template sniffing cannot separate the two and would fall
back to the wrong parser. With qwen3_coder named, tool calls arrive as real
tool_calls on the OpenAI response.
This is the bf16 checkpoint, roughly 57 GB of weights, which is what the
engine was gated on. Being bf16 rather than NVFP4 it does not need Blackwell
on its own account, but LocalAI's CUDA images for this backend are currently
built for Blackwell-family GPUs only, so on an older card use the CPU build.
Repository: localaiLicense: apache-2.0
qwen3-4b-vllm-cpp
Qwen3-4B on vllm.cpp, in bf16. The small end of the engine's gated dense
family, which reaches parity with vLLM on every axis at concurrency 1.
bf16 rather than NVFP4 on purpose: this is the entry that runs where the
flagship NVFP4 checkpoints cannot, including Apple Silicon via Metal, Vulkan
and plain CPU. Roughly 8 GB of weights, plus about 4.5 GB of KV cache at the
context configured here.
Tool calling and the thinking split are parsed inside the engine.
Repository: localaiLicense: apache-2.0
qwen3-0.6b-vllm-cpp
Qwen3-0.6B on vllm.cpp, in bf16. Roughly 1.4 GB of weights, which makes it
the cheapest way to confirm a vllm-cpp install actually serves before
committing disk and memory to one of the large checkpoints.
It runs anywhere the backend does, CPU included, and it is a real chat model
rather than a stub, so tool calling and the thinking split can be exercised
on it too.
Repository: localaiLicense: apache-2.0
minimax-h3-fl2va-unpruned-q4
MiniMax-H3 served by vllm.cpp, LocalAI's own C++ port of vLLM. It generates
video AND audio jointly, so a clip comes back as an MP4 with a soundtrack.
Ask for speech in the prompt and the model lip-syncs it.
Runtime: vllm.cpp (backend vllm-cpp). Partition: fl2va. DiT layout:
unpruned. Format: Q4_K_M.
This entry is the FL2VA partition, which serves text-to-video (t2va) and first/last-frame conditioning (fl2va). It refuses reference conditioning: for that use the matching ref2va entry. The DiT is the original time-embedder DiT (~19.9 GB).
Render status: verified render (end to end on the vllm.cpp model page).
Q3_K_M of the same DiT is verified BAD and is not offered.
The vllm.cpp page records its renders with the DiT expanded to bf16. This
entry does not expand it and keeps the quantised weights resident, so a
render in that mode is not covered by the page.
The weights are one DiT, the Qwen3-VL-32B encoder (~14.6 GB), the video VAE
(~10.4 GB), the audio VAE and the tokenizer, plus the two VAE configs that
carry the latent statistics. The default canvas is 1344x768 at 124 frames
and 24 fps, about 5.2 seconds. Generation is slow: roughly 176 s per denoise
step at that canvas on a 20-SM device. Muxing the finished frames needs
ffmpeg on the host.
Repository: localaiLicense: other
minimax-h3-fl2va-q4
MiniMax-H3 served by vllm.cpp, LocalAI's own C++ port of vLLM. It generates
video AND audio jointly, so a clip comes back as an MP4 with a soundtrack.
Ask for speech in the prompt and the model lip-syncs it.
Runtime: vllm.cpp (backend vllm-cpp). Partition: fl2va. DiT layout:
unpruned. Format: Q4_K_M.
This entry is the FL2VA partition, which serves text-to-video (t2va) and first/last-frame conditioning (fl2va). It refuses reference conditioning: for that use the matching ref2va entry. The DiT is the original time-embedder DiT (~19.9 GB).
Render status: verified render (end to end on the vllm.cpp model page).
Q3_K_M of the same DiT is verified BAD and is not offered.
The vllm.cpp page records its renders with the DiT expanded to bf16. This
entry does not expand it and keeps the quantised weights resident, so a
render in that mode is not covered by the page.
The weights are one DiT, the Qwen3-VL-32B encoder (~14.6 GB), the video VAE
(~10.4 GB), the audio VAE and the tokenizer, plus the two VAE configs that
carry the latent statistics. The default canvas is 1344x768 at 124 frames
and 24 fps, about 5.2 seconds. Generation is slow: roughly 176 s per denoise
step at that canvas on a 20-SM device. Muxing the finished frames needs
ffmpeg on the host.
Repository: localaiLicense: other
minimax-h3-ref2va-unpruned-q4
MiniMax-H3 served by vllm.cpp, LocalAI's own C++ port of vLLM. It generates
video AND audio jointly, so a clip comes back as an MP4 with a soundtrack.
Ask for speech in the prompt and the model lip-syncs it.
Runtime: vllm.cpp (backend vllm-cpp). Partition: ref2va. DiT layout:
unpruned. Format: Q4_K_M.
This entry is the Ref2VA partition, which takes REFERENCE conditioning (reference images, clips and audio) so the subject or voice carries into the clip. It refuses plain text-to-video and first/last-frame requests: for those use the matching fl2va entry. The DiT is the original time-embedder DiT (~19.9 GB).
Render status: verified render (end to end on the vllm.cpp model page).
Do NOT use the NVFP4 Ref2VA weights: they render a multicolour patch grid.
Q3_K_M of the same DiT is verified BAD and is not offered.
The vllm.cpp page records its renders with the DiT expanded to bf16. This
entry does not expand it and keeps the quantised weights resident, so a
render in that mode is not covered by the page.
The weights are one DiT, the Qwen3-VL-32B encoder (~14.6 GB), the video VAE
(~10.4 GB), the audio VAE and the tokenizer, plus the two VAE configs that
carry the latent statistics. The default canvas is 1344x768 at 124 frames
and 24 fps, about 5.2 seconds. Generation is slow: roughly 176 s per denoise
step at that canvas on a 20-SM device. Muxing the finished frames needs
ffmpeg on the host.
Repository: localaiLicense: other
minimax-h3-ref2va-q4
MiniMax-H3 served by vllm.cpp, LocalAI's own C++ port of vLLM. It generates
video AND audio jointly, so a clip comes back as an MP4 with a soundtrack.
Ask for speech in the prompt and the model lip-syncs it.
Runtime: vllm.cpp (backend vllm-cpp). Partition: ref2va. DiT layout:
unpruned. Format: Q4_K_M.
This entry is the Ref2VA partition, which takes REFERENCE conditioning (reference images, clips and audio) so the subject or voice carries into the clip. It refuses plain text-to-video and first/last-frame requests: for those use the matching fl2va entry. The DiT is the original time-embedder DiT (~19.9 GB).
Render status: verified render (end to end on the vllm.cpp model page).
Do NOT use the NVFP4 Ref2VA weights: they render a multicolour patch grid.
Q3_K_M of the same DiT is verified BAD and is not offered.
The vllm.cpp page records its renders with the DiT expanded to bf16. This
entry does not expand it and keeps the quantised weights resident, so a
render in that mode is not covered by the page.
The weights are one DiT, the Qwen3-VL-32B encoder (~14.6 GB), the video VAE
(~10.4 GB), the audio VAE and the tokenizer, plus the two VAE configs that
carry the latent statistics. The default canvas is 1344x768 at 124 frames
and 24 fps, about 5.2 seconds. Generation is slow: roughly 176 s per denoise
step at that canvas on a 20-SM device. Muxing the finished frames needs
ffmpeg on the host.
Repository: localaiLicense: other
minimax-h3-fl2va-pruned-q8
MiniMax-H3 served by vllm.cpp, LocalAI's own C++ port of vLLM. It generates
video AND audio jointly, so a clip comes back as an MP4 with a soundtrack.
Ask for speech in the prompt and the model lip-syncs it.
Runtime: vllm.cpp (backend vllm-cpp). Partition: fl2va. DiT layout:
pruned. Format: Q8_0.
This entry is the FL2VA partition, which serves text-to-video (t2va) and first/last-frame conditioning (fl2va). It refuses reference conditioning: for that use the matching ref2va entry. The DiT is the pruned DiT (AdaLN curve table in place of the per-block modulation projection; about 20B instead of 33B parameters) (~21.4 GB).
Render status: loads and renders (vllm.cpp model page, pruned Q8_0 A/B).
The vllm.cpp page records its renders with the DiT expanded to bf16. This
entry does not expand it and keeps the quantised weights resident, so a
render in that mode is not covered by the page.
The weights are one DiT, the Qwen3-VL-32B encoder (~14.6 GB), the video VAE
(~10.4 GB), the audio VAE and the tokenizer, plus the two VAE configs that
carry the latent statistics. The default canvas is 1344x768 at 124 frames
and 24 fps, about 5.2 seconds. Generation is slow: roughly 176 s per denoise
step at that canvas on a 20-SM device. Muxing the finished frames needs
ffmpeg on the host.
Repository: localaiLicense: other
Page 1