LOCAL MODEL SUPPORT, WITH SOURCES
Does it run local yet?
Live support status for every new open model across llama.cpp, Ollama, LM Studio, MLX and vLLM.
12 models · public-source snapshot · observed 2026-10-19 10:00 UTC
Runtime support
12 models
Architecture and upload listings are evidence only; exact model execution is unverified.
Scroll within the comparison to see all runtimes.
| Model | llama.cpp | Ollama | LM Studio | MLX / mlx-lm | vLLM |
|---|---|---|---|---|---|
| DeepSeek-V4.1-Flash-1017deepseek-ai · created 17 Oct 2026observed 19 Oct 2026 | UnknownDeepSeek-V4.1-Flash-1017 · llama.cpp Recorded status: Unknown Unknown — no verified support evidence. Checked the exact model config and the runtime architecture registry; no matching declared architecture was found. This is unknown, not evidence of incompatibility. Checked: 2026-10-19 02:17 UTC | UnknownDeepSeek-V4.1-Flash-1017 · Ollama Recorded status: Unknown Unknown — no verified support evidence. Ollama exposes the exact deepseek-v4.1-flash:1017-cloud cloud variant; local model availability remains unknown. Checked: 2026-10-19 10:01 UTC | UnknownDeepSeek-V4.1-Flash-1017 · LM Studio Recorded status: Unknown Unknown — no verified support evidence. Fresh LM Studio page identifies this model as cloud-only; local model availability remains unknown. Checked: 2026-10-19 10:01 UTC | Upload availableDeepSeek-V4.1-Flash-1017 · MLX / mlx-lm Recorded status: Upload available Fresh exact base_model-tagged public repository contains weight files in a format used by mlx_lm. This establishes upload availability only; model compatibility and execution were not tested. Checked: 2026-10-19 10:01 UTC | Architecture listedDeepSeek-V4.1-Flash-1017 · vLLM Recorded status: Architecture listed The model config architecture is listed in this runtime registry. This is family/architecture evidence only; exact checkpoint loading, model-specific configuration, hardware, and generation were not tested. Checked: 2026-10-19 02:17 UTC |
| NVIDIA-Nemotron-Labs-Dory1.0-4B-BF16nvidia · created 16 Oct 2026observed 19 Oct 2026 | UnknownNVIDIA-Nemotron-Labs-Dory1.0-4B-BF16 · llama.cpp Recorded status: Unknown Unknown — no verified support evidence. Checked the exact model config and the runtime architecture registry; no matching declared architecture was found. This is unknown, not evidence of incompatibility. Checked: 2026-10-19 10:00 UTC | Unknown v0.40.6NVIDIA-Nemotron-Labs-Dory1.0-4B-BF16 · Ollama Recorded status: Unknown Unknown — no verified support evidence. The exact model and runtime-specific sources were checked; no positive model-level evidence was established. Version: v0.40.6 Checked: 2026-10-19 10:00 UTC | UnknownNVIDIA-Nemotron-Labs-Dory1.0-4B-BF16 · LM Studio Recorded status: Unknown Unknown — no verified support evidence. The exact model and runtime-specific sources were checked; no positive model-level evidence was established. Checked: 2026-10-19 10:00 UTC | UnknownNVIDIA-Nemotron-Labs-Dory1.0-4B-BF16 · MLX / mlx-lm Recorded status: Unknown Unknown — no verified support evidence. Checked the exact model config and the runtime architecture registry; no matching declared architecture was found. This is unknown, not evidence of incompatibility. Checked: 2026-10-19 10:00 UTC | UnknownNVIDIA-Nemotron-Labs-Dory1.0-4B-BF16 · vLLM Recorded status: Unknown Unknown — no verified support evidence. Checked the exact model config and the runtime architecture registry; no matching declared architecture was found. This is unknown, not evidence of incompatibility. Checked: 2026-10-19 10:00 UTC |
| Hy-Memory-30B-A3Btencent · created 15 Oct 2026observed 19 Oct 2026 | Architecture listedHy-Memory-30B-A3B · llama.cpp Recorded status: Architecture listed The model config architecture is listed in this runtime registry. This is family/architecture evidence only; exact checkpoint loading, model-specific configuration, hardware, and generation were not tested. Checked: 2026-10-19 02:17 UTC | UnknownHy-Memory-30B-A3B · Ollama Recorded status: Unknown Unknown — no verified support evidence. The exact-name Ollama search returned no model. A community GGUF exists on Hugging Face, but that is not an Ollama catalog/upload entry. No compatibility conclusion is implied. Checked: 2026-10-19 10:01 UTC | UnknownHy-Memory-30B-A3B · LM Studio Recorded status: Unknown Unknown — no verified support evidence. Targeted public search of the lmstudio-community namespace produced no exact matching model with base_model metadata; absence is not evidence of runtime incompatibility. Checked: 2026-10-19 10:01 UTC | UnknownHy-Memory-30B-A3B · MLX / mlx-lm Recorded status: Unknown Unknown — no verified support evidence. Targeted public search of mlx-community yielded no exact base_model match with MLX file metadata; absence does not mean unsupported. Checked: 2026-10-19 10:01 UTC | Architecture listedHy-Memory-30B-A3B · vLLM Recorded status: Architecture listed The model config architecture is listed in this runtime registry. This is family/architecture evidence only; exact checkpoint loading, model-specific configuration, hardware, and generation were not tested. Checked: 2026-10-19 02:17 UTC |
| LFM2.5-24B-A2BLiquidAI · created 12 Oct 2026observed 19 Oct 2026 | Architecture listedLFM2.5-24B-A2B · llama.cpp Recorded status: Architecture listed The model config architecture is listed in this runtime registry. This is family/architecture evidence only; exact checkpoint loading, model-specific configuration, hardware, and generation were not tested. Checked: 2026-10-19 02:17 UTC | ReleasedLFM2.5-24B-A2B · Ollama Recorded status: In release In release. Fresh Ollama curated library tags page exposes this exact local model variant; catalog availability only, not hardware-tested support. Checked: 2026-10-19 10:01 UTC | UnknownLFM2.5-24B-A2B · LM Studio Recorded status: Unknown Unknown — no verified support evidence. Targeted public search of the lmstudio-community namespace produced no exact matching model with base_model metadata; absence is not evidence of runtime incompatibility. Checked: 2026-10-19 10:01 UTC | Upload availableLFM2.5-24B-A2B · MLX / mlx-lm Recorded status: Upload available Fresh exact base_model-tagged public repository contains weight files in a format used by mlx_lm. This establishes upload availability only; model compatibility and execution were not tested. Checked: 2026-10-19 10:01 UTC | Architecture listedLFM2.5-24B-A2B · vLLM Recorded status: Architecture listed The model config architecture is listed in this runtime registry. This is family/architecture evidence only; exact checkpoint loading, model-specific configuration, hardware, and generation were not tested. Checked: 2026-10-19 02:17 UTC |
| NVIDIA-Nemotron-3.5-Bolt-74B-A9B-24GBnvidia · created 07 Oct 2026observed 19 Oct 2026 | Projector listedNVIDIA-Nemotron-3.5-Bolt-74B-A9B-24GB · llama.cpp Recorded status: Projector component listed The exact config architecture appears in llama.cpp MMPROJ_MODEL_MAP only. Projector conversion is listed; full text model conversion and runtime support are unverified. Checked: 2026-10-19 02:18 UTC | ReleasedNVIDIA-Nemotron-3.5-Bolt-74B-A9B-24GB · Ollama Recorded status: In release In release. Fresh Ollama curated library tags page exposes this exact local model variant; catalog availability only, not hardware-tested support. Checked: 2026-10-19 10:01 UTC | Upload availableNVIDIA-Nemotron-3.5-Bolt-74B-A9B-24GB · LM Studio Recorded status: Upload available Fresh public repository README contains the exact model id or full title and the repository contains GGUF weight files. The repository omits base_model metadata, so this supports community upload availability only; no exact linked catalog release was verified. Checked: 2026-10-19 10:01 UTC | UnknownNVIDIA-Nemotron-3.5-Bolt-74B-A9B-24GB · MLX / mlx-lm Recorded status: Unknown Unknown — no verified support evidence. Targeted public search of mlx-community yielded no exact base_model match with MLX file metadata; absence does not mean unsupported. Checked: 2026-10-19 10:01 UTC | Architecture listedNVIDIA-Nemotron-3.5-Bolt-74B-A9B-24GB · vLLM Recorded status: Architecture listed The model config architecture is listed in this runtime registry. This is family/architecture evidence only; exact checkpoint loading, model-specific configuration, hardware, and generation were not tested. Checked: 2026-10-19 02:18 UTC |
| MiniCPM-V-4.7-35B-A3Bopenbmb · created 07 Oct 2026observed 19 Oct 2026 | Architecture listedMiniCPM-V-4.7-35B-A3B · llama.cpp Recorded status: Architecture listed The model config architecture is listed in this runtime registry. This is family/architecture evidence only; exact checkpoint loading, model-specific configuration, hardware, and generation were not tested. Checked: 2026-10-19 02:17 UTC | Upload availableMiniCPM-V-4.7-35B-A3B · Ollama Recorded status: Upload available Fresh Ollama publisher tags page exposes this exact local upload variant; availability only, not hardware-tested support. Checked: 2026-10-19 10:01 UTC | UnknownMiniCPM-V-4.7-35B-A3B · LM Studio Recorded status: Unknown Unknown — no verified support evidence. Targeted public search of the lmstudio-community namespace produced no exact matching model with base_model metadata; absence is not evidence of runtime incompatibility. Checked: 2026-10-19 10:01 UTC | Upload availableMiniCPM-V-4.7-35B-A3B · MLX / mlx-lm Recorded status: Upload available Fresh exact base_model-tagged repository contains MLX community weight files and is tagged mlx-vlm. This establishes upload availability only; mlx-lm compatibility and execution were not tested. Checked: 2026-10-19 10:01 UTC | MergedMiniCPM-V-4.7-35B-A3B · vLLM Recorded status: Merged · release unverified Merged · release unverified. Official MiniCPM-V 4.7 deployment guide links this exact model ID and PR #58674; the support PR is merged, while an official release is not established. Checked: 2026-10-19 10:00 UTC |
| MiniCPM-V-4.7-1Bopenbmb · created 06 Oct 2026observed 19 Oct 2026 | Architecture listedMiniCPM-V-4.7-1B · llama.cpp Recorded status: Architecture listed The model config architecture is listed in this runtime registry. This is family/architecture evidence only; exact checkpoint loading, model-specific configuration, hardware, and generation were not tested. Checked: 2026-10-19 02:18 UTC | Upload availableMiniCPM-V-4.7-1B · Ollama Recorded status: Upload available Fresh Ollama publisher tags page exposes this exact local upload variant; availability only, not hardware-tested support. Checked: 2026-10-19 10:01 UTC | UnknownMiniCPM-V-4.7-1B · LM Studio Recorded status: Unknown Unknown — no verified support evidence. Targeted public search of the lmstudio-community namespace produced no exact matching model with base_model metadata; absence is not evidence of runtime incompatibility. Checked: 2026-10-19 10:01 UTC | Upload availableMiniCPM-V-4.7-1B · MLX / mlx-lm Recorded status: Upload available Fresh exact base_model-tagged repository contains MLX community weight files and is tagged mlx-vlm. This establishes upload availability only; mlx-lm compatibility and execution were not tested. Checked: 2026-10-19 10:01 UTC | MergedMiniCPM-V-4.7-1B · vLLM Recorded status: Merged · release unverified Merged · release unverified. Official MiniCPM-V 4.7 deployment guide links this exact model ID and PR #58674; the support PR is merged, while an official release is not established. Checked: 2026-10-19 10:00 UTC |
| Muse-Spark-Open-816B-A42Bmeta-models · created 15 Sep 2026observed 19 Oct 2026 | Architecture listedMuse-Spark-Open-816B-A42B · llama.cpp Recorded status: Architecture listed The model config architecture is listed in this runtime registry. This is family/architecture evidence only; exact checkpoint loading, model-specific configuration, hardware, and generation were not tested. Checked: 2026-10-19 02:17 UTC | UnknownMuse-Spark-Open-816B-A42B · Ollama Recorded status: Unknown Unknown — no verified support evidence. The exact-name Ollama search returned no matching model entry. This is a public-search gap, not evidence of incompatibility. Checked: 2026-10-19 10:01 UTC | UnknownMuse-Spark-Open-816B-A42B · LM Studio Recorded status: Unknown Unknown — no verified support evidence. Targeted lmstudio-community API search returned no exact base_model upload. The local identity remains unverified; a similarly named cloud entry does not prove the exact model is cloud-only. Checked: 2026-10-19 10:01 UTC | Upload availableMuse-Spark-Open-816B-A42B · MLX / mlx-lm Recorded status: Upload available Fresh exact base_model-tagged public repository contains weight files in a format used by mlx_lm. This establishes upload availability only; model compatibility and execution were not tested. Checked: 2026-10-19 10:01 UTC | UnknownMuse-Spark-Open-816B-A42B · vLLM Recorded status: Unknown Unknown — no verified support evidence. Checked the exact model config and the runtime architecture registry; no matching declared architecture was found. This is unknown, not evidence of incompatibility. Checked: 2026-10-19 02:17 UTC |
| DeepSeek-V4.1-Flashdeepseek-ai · created 10 Sep 2026observed 19 Oct 2026 | UnknownDeepSeek-V4.1-Flash · llama.cpp Recorded status: Unknown Unknown — no verified support evidence. Checked the exact model config and the runtime architecture registry; no matching declared architecture was found. This is unknown, not evidence of incompatibility. Checked: 2026-10-19 02:17 UTC | UnknownDeepSeek-V4.1-Flash · Ollama Recorded status: Unknown Unknown — no verified support evidence. Ollama exposes the exact deepseek-v4.1-flash:cloud cloud variant; local model availability remains unknown. Checked: 2026-10-19 10:01 UTC | UnknownDeepSeek-V4.1-Flash · LM Studio Recorded status: Unknown Unknown — no verified support evidence. Fresh LM Studio page identifies this model as cloud-only; local model availability remains unknown. Checked: 2026-10-19 10:01 UTC | Architecture listedDeepSeek-V4.1-Flash · MLX / mlx-lm Recorded status: Architecture listed The model config architecture is listed in this runtime registry. This is family/architecture evidence only; exact checkpoint loading, model-specific configuration, hardware, and generation were not tested. Checked: 2026-10-19 02:17 UTC | MergedDeepSeek-V4.1-Flash · vLLM Recorded status: Merged · release unverified Merged · release unverified. Model-specific vLLM support PR; a merged PR does not establish an official release. Checked: 2026-10-19 10:00 UTC |
| Qwen3.8-Flash-NextQwen · created 24 Aug 2026observed 19 Oct 2026 | Architecture listedQwen3.8-Flash-Next · llama.cpp Recorded status: Architecture listed The model config architecture is listed in this runtime registry. This is family/architecture evidence only; exact checkpoint loading, model-specific configuration, hardware, and generation were not tested. Checked: 2026-10-19 02:17 UTC | ReleasedQwen3.8-Flash-Next · Ollama Recorded status: In release In release. Ollama public library page exposes this exact size variant; catalog availability only, not hardware-tested support. Checked: 2026-10-19 10:00 UTC | ReleasedQwen3.8-Flash-Next · LM Studio Recorded status: In release In release. Fresh LM Studio local catalog card identifies this exact model, and its exact base_model-tagged GGUF repository contains weight files. Execution remains untested. Checked: 2026-10-19 10:01 UTC | Upload availableQwen3.8-Flash-Next · MLX / mlx-lm Recorded status: Upload available A current public MLX upload with an exact base_model match was observed. Upload availability only; this model was not loaded or hardware-tested. Checked: 2026-10-19 10:00 UTC | Architecture listedQwen3.8-Flash-Next · vLLM Recorded status: Architecture listed The model config architecture is listed in this runtime registry. This is family/architecture evidence only; exact checkpoint loading, model-specific configuration, hardware, and generation were not tested. Checked: 2026-10-19 02:17 UTC |
| Qwen3.8-27BQwen · created 05 Aug 2026observed 19 Oct 2026 | Architecture listedQwen3.8-27B · llama.cpp Recorded status: Architecture listed The model config architecture is listed in this runtime registry. This is family/architecture evidence only; exact checkpoint loading, model-specific configuration, hardware, and generation were not tested. Checked: 2026-10-19 02:17 UTC | ReleasedQwen3.8-27B · Ollama Recorded status: In release In release. Fresh Ollama curated library tags page exposes this exact local model variant; catalog availability only, not hardware-tested support. Checked: 2026-10-19 10:01 UTC | ReleasedQwen3.8-27B · LM Studio Recorded status: In release In release. Fresh LM Studio local catalog card identifies this exact model, and its exact base_model-tagged GGUF repository contains weight files. Execution remains untested. Checked: 2026-10-19 10:01 UTC | Upload availableQwen3.8-27B · MLX / mlx-lm Recorded status: Upload available Fresh exact base_model-tagged public repository contains weight files in a format used by mlx_lm. This establishes upload availability only; model compatibility and execution were not tested. Checked: 2026-10-19 10:01 UTC | Architecture listedQwen3.8-27B · vLLM Recorded status: Architecture listed The model config architecture is listed in this runtime registry. This is family/architecture evidence only; exact checkpoint loading, model-specific configuration, hardware, and generation were not tested. Checked: 2026-10-19 02:17 UTC |
| Unlimited-OCRbaidu · created 19 Jun 2026observed 19 Oct 2026 | Architecture listedUnlimited-OCR · llama.cpp Recorded status: Architecture listed The model config architecture is listed in this runtime registry. This is family/architecture evidence only; exact checkpoint loading, model-specific configuration, hardware, and generation were not tested. Checked: 2026-10-19 02:18 UTC | Upload availableUnlimited-OCR · Ollama Recorded status: Upload available Fresh Ollama publisher tags page exposes this exact local upload variant; availability only, not hardware-tested support. Checked: 2026-10-19 10:01 UTC | UnknownUnlimited-OCR · LM Studio Recorded status: Unknown Unknown — no verified support evidence. Targeted public search of the lmstudio-community namespace produced no exact matching model with base_model metadata; absence is not evidence of runtime incompatibility. Checked: 2026-10-19 10:01 UTC | Upload availableUnlimited-OCR · MLX / mlx-lm Recorded status: Upload available Fresh exact base_model-tagged repository contains MLX community weight files and is tagged mlx-vlm. This establishes upload availability only; mlx-lm compatibility and execution were not tested. Checked: 2026-10-19 10:01 UTC | Architecture listedUnlimited-OCR · vLLM Recorded status: Architecture listed The model config architecture is listed in this runtime registry. This is family/architecture evidence only; exact checkpoint loading, model-specific configuration, hardware, and generation were not tested. Checked: 2026-10-19 02:18 UTC |
No models match these filters.
READING THE EVIDENCE
Availability has layers.
“Released” records an exact model entry in a runtime catalog or release source. “Architecture listed” is a registry match for an architecture, not the model’s weights. “Projector component listed” confirms only a projector conversion component. “Upload available” records a matching community upload, not verified runtime integration. “Merged” means the change merged; release remains unverified. “Not yet” is an explicit unsupported status in the evidence, not a prediction about future support. Unknown means evidence is missing or inconclusive.
Memory suggestions use observed complete GGUF quant files and a headroom estimate where the catalog has a qualifying recommendation. They do not predict runtime compatibility, speed, context length, or usable context length. Stale evidence is labeled, and an empty curated issue list does not mean there are no bugs.
HARDWARE NOTES
Tested on DGX Spark / Mac Studio: coming soon
No hardware benchmark measurements are available in this snapshot.