Youtu-Parsing-Omni on local hardware
Estimated comfort for repository coding with Aider on a MacBook Pro M4 Pro 48GB.
Youtu-Parsing-Omni
tencent · other · 1,927 downloads
VisionReasoning / thinking mode
- Parameters
- 5.3B exact
- Context window
- 1M tokens
- Attention
- Multi-head latent attention (compressed KV)
- Layers · KV heads · head dim
- 40 · 8 · 64
- KV cache
- 45 KB/token · 1.4 GB at 32K
- Tool calling
- basic (from chat template)
- Thinking
- good (from chat template)
- Vision
- good (from config)
- Commercial use
- check cardCheck the model card
- Import confidence
- medium · capabilities estimated
Weights by quantization
Q3_K_M
2.5 GB
computed · low confidence
Q4_K_M
3.1 GB
computed · low confidence
Q5_K_M
3.6 GB
computed · low confidence
Q6_K
4.2 GB
computed · low confidence
Q8_0
5.4 GB
computed · low confidence
F16 / BF16
9.9 GB
safetensors file size
- • Assumes GGUF / MLX conversions of this model exist (they do for most popular models). Sizes are computed from bits per weight.
- • vision: good, inferred from config (low confidence).
- • Capability ratings are estimated from model size and name — this model hasn't been assessed by hand.
- • The chat template doesn't mention tools, so tool calling is assumed to be limited.
- • Commercial use unknown: Check the model card.
Youtu-Parsing-Omni — 8-bit
Borderline for your workload
Speed
25–35 tok/s
Comfortable
Memory left
29 GB
of 48 GB
First word in
≈30 s
Adequate
All the details
- Your setup
- MacBook Pro M4 Pro 48 GB
- Workload
- Repository coding with Aider
- Runtime → tool
- LM Studio → Aider
- Memory used
- 7.9 GB
- Context
- 32K · up to 128K practical
- Tool support
- Official
- Use-case fit
- Poorly suited
Why?
- Model fits with approximately 29 GB of memory headroom.
- LM Studio supports hardware acceleration on this machine (Metal (Apple GPU)).
- Aider can connect to LM Studio (officially supported).
- Generation (25–35 tok/s) should be appropriate for interactive coding.
- 32K working context is reasonable.
- Ingesting a prompt of 14K tokens takes ≈30 s, and each step ≈45 s — noticeable waits.
- Weak coding capability for interactive coding.
Why “Borderline” rather than “Acceptable”?
- Workload suitability: Weak coding capability for interactive coding.
- Prompt processing: Ingesting the first prompt (14K tokens) takes ≈30 s; each of the ~4 follow-up turns takes ≈45 s (≈3 min per task).
How sure are we?
Medium confidenceEstimated from bandwidth and compute, calibrated against 2 verified benchmarks of other models on this same chip (measured with a different engine, so calibration is partial).
- Calibrated with verified benchmarks of other models on this chip.
Medium confidence