Over recent months, we have intensified our focus on developing foundation models that deliver exceptional utility and performance.
WeMM-Video-2B
tencent·Instruct·released 17 Nov 2026
hf download tencent/WeMM-Video-2BWeMM-Video-2B is the dense 2B checkpoint in the WeMM-Video family. WeMM-Video is a Chinese–English bilingual family for joint audio-visual understanding and reasoning. It accepts text, images, audio, and video and generates text responses.
How much VRAM does WeMM-Video-2B need?
At Q4_K_M WeMM-Video-2B needs about 2.8 GB with an 8K context (2.8 GB at Q4, 7.3 GB at full precision). Pick your GPU to see what fits and how fast it runs.
Needs: ≈2.8 GB · Q4_K_M · 8K context — Pick a GPU to check the fit.
Advanced: context 8K · KV cache 16-bit
Smallest GPU for each quant and context (KV cache 16-bit)
| Quantisation | 8K | 16K | 32K | 64K | 128K | 256K |
|---|---|---|---|---|---|---|
| Q3_K_M | ||||||
| Q4_K_M | ||||||
| Q5_K_M | ||||||
| Q6_K | ||||||
| Q8_0 | ||||||
| FP16 |
≤16 GB card24–32 GB48–96 GBmore
| Quantisation | Total |
|---|---|
| Q2_KHeavy lossSmallest, noticeably worse quality | ≈1.9 GB |
| Q3_K_MNoticeable lossSmall, some quality loss | ≈2.4 GB |
| Q4_K_MrecommendedSmall lossRecommended default for local use | ≈2.8 GB |
| Q5_K_MNear-losslessNear-lossless for most tasks | ≈3.2 GB |
| Q6_KNear-losslessPractically lossless | ≈3.5 GB |
| Q8_0LosslessLossless in practice, twice the size of Q4 | ≈4.3 GB |
| FP16 / BF16LosslessFull precision, as released | ≈7.3 GB |
Estimates for 8K tokens of context: weights + KV cache + runtime overhead. Speed assumes memory-bound decoding at your GPU’s bandwidth. Pick a GPU above and the fit column checks every row against it.
How to pick a quant
Take the biggest quant that fits with room left for context. Q4_K_M is the usual sweet spot. Below Q4 quality drops fast, above Q6 you mostly pay memory for nothing. Big models (70B+, large MoE) tolerate low quants better than small ones.
- LosslessF16, Q8_0
- Near-losslessQ6_K, Q5_K_M
- Small lossQ5_K_S, Q4_K_M, IQ4_NL, IQ4_XS, Q4_K_S
- Noticeable lossQ4_0, Q3_K_L, Q3_K_M, IQ3_M
- Heavy lossIQ3_XXS, Q3_K_S, IQ2_M, Q2_K
Typical quality loss, not measured per model. Tiers come from KL divergence against FP16 on Llama 3 8B: llama.cpp perplexity README: LLaMA 3 8B scoreboard (KLD and ΔPPL vs FP16, Wikitext-2); llama.cpp quantize tool: ΔPPL per quant type @ Llama-3-8B
How much context fits?
At 8K context, Q4_K_M needs ≈2.8 GB(1.9 GB weights + 0.4 GB KV cache + 0.6 GB overhead). Pick a GPU above to see how much context fits.
Weights + overheadKV cache
Can the RTX 4060 run WeMM-Video-2B? →
Speed vs quality per quant
Smaller quants run faster and fit in less memory, but lose quality.
Fits in VRAMTight fitCPU offload✕ won't runQuality is typical KL divergence vs FP16, not measured for this model.
Real-world speed of WeMM-Video-2B
Tokens per second measured by people running it at home. A GPU shows a median once 3 reports agree on the same setup. Reports far from our estimate wait for a review. Measure yours with llama-bench →
No medians yet. Be one of the first 3 to report a setup.
All reports
Run WeMM-Video-2B locally
Quantised versions and downloads
| Repository | Format | Quant | Size |
|---|---|---|---|
| tencent/WeMM-Video-2B-NVFP4 | Safetensors | NVFP4 | — |
| tencent/WeMM-Video-2B-FP8 | Safetensors | FP8 | — |
| tencent/WeMM-Video-2B-AWQ | AWQ | AWQ | — |
About WeMM-Video-2B
WeMM-Video-2B is the dense 2B checkpoint in the WeMM-Video family. WeMM-Video is a Chinese–English bilingual family for joint audio-visual understanding and reasoning. It accepts text, images, audio, and video and generates text responses. • Weights · BF16 Safetensors · 6.12 GiB of weight files • Quantized variants · FP8 Block128 · AWQ W4A16 · GGUF Q4_K_M · NVFP4 W4A16 WeMM-Video can combine video frames with the original soundtrack to describe events, transcribe speech, and reason about what is seen and heard. It also accepts silent video, standalone audio, images, and text, with responses in Chinese or English. Vision and audio encoders turn media into tokens that the language model processes alongside text. Video frames carry timestamps. The family was pretrained on approximately 1.2 trillion multimodal tokens, then post-trained with multi-teacher OPD, anti-repetition MrDPO, and specialist reinforcement learning. See the project overview for family details. Set up the Transformers environment. From the WeMM-Video repository root, run the example on an MP4 with a soundtrack. The script loads the model and processor from the Hub and processes both the frames and the original audio track. Use --revision to pin the model and processor to a Hub commit. Size GPU memory for the full weight files, media inputs, and generation cache.
Full model card on Hugging Face →Alternatives to WeMM-Video-2B
All alternatives →Over recent months, we have intensified our focus on developing foundation models that deliver exceptional utility and performance.
LFM2.5-VL-3B is a multimodal variant of LFM2.5, a family of hybrid models designed for on-device deployment. It builds on LFM2-VL-3B with further mid- and post-training.
d1-3B is a 3B parameter decision model built on LFM2.5-VL-3B. You give it a state (text, JSON, images, or a mix) and a set of questions. It returns calibrated, typed answers in one forward pass with zero output tokens.
Meet Qwen3-VL — the most powerful vision-language model in the Qwen series to date.
Muse Glimmer is a 30-billion-parameter causal language model with a dedicated perception encoder, distilled from Muse Spark and purpose-built for autonomous agentic tasks on consumer hardware.
WeMM-Video-2B FAQ
- How much VRAM does WeMM-Video-2B need?
- About 2.8 GB at Q4_K_M with an 8K context. Q8_0 needs about 4.3 GB and full precision about 7.3 GB. Longer contexts need more memory for the KV cache.
- Can WeMM-Video-2B run on the RTX 4060?
- Yes. It fits at FP16 / BF16 (about 7.3 GB), at an estimated 23 tokens/s. It fits comfortably on the Apple M4 (16 GB), Radeon RX 6700 XT, RTX 3060 12GB.
- Can I use WeMM-Video-2B commercially?
- Yes. It is released under Apache 2.0. Permissive: use, modify and sell freely; keep the notice and licence text.
- What is the context length of WeMM-Video-2B?
- 262,144 tokens (256K). The full window needs far more memory than an 8K context.
Rate this model
Data from Hugging Face, the model card and public leaderboards. Memory and speed figures are estimates; real numbers depend on the runtime, context length and drivers. Always check the licence yourself before commercial use.
0 comments