AesCode-8B on local hardware
Estimated comfort for repository coding with Aider on a MacBook Pro M4 Pro 48GB.
AesCode-8BOpen source
microsoft · apache-2.0 · 950 downloads
Vision
- Parameters
- 8.8B exact
- Context window
- 256K tokens
- Attention
- Standard (full attention)
- Layers · KV heads · head dim
- 36 · 8 · 128
- KV cache
- 144 KB/token · 4.5 GB at 32K
- Tool calling
- good (from chat template)
- Thinking
- basic (from chat template)
- Vision
- good (from config)
- Commercial use
- allowedRetain the required license and attribution notices
- Import confidence
- medium · capabilities estimated
Weights by quantization
Q3_K_M
3.8 GB
gguf file size
Q4_K_M
4.9 GB
gguf file size
Q5_K_M
5.8 GB
gguf file size
Q6_K
6.6 GB
gguf file size
Q8_0
8.1 GB
gguf file size
F16 / BF16
15 GB
gguf file size
- • vision: good, inferred from config (low confidence).
- • Capability ratings are estimated from model size and name — this model hasn't been assessed by hand.
- • GGUF sizes come from community conversions matched by model name; equivalence and runtime support are not verified. MLX conversion availability is unverified.
AesCode-8B — 6-bit
Acceptable for your workload
Speed
17–25 tok/s
Adequate
Memory left
24 GB
of 48 GB
First word in
≈50 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
- 12 GB
- Context
- 32K · up to 64K practical
- Tool support
- Official
- Use-case fit
- Partly suited
Why?
- Model fits with approximately 24 GB of memory headroom.
- LM Studio supports hardware acceleration on this machine (Metal (Apple GPU)).
- Aider can connect to LM Studio (officially supported).
- Generation (17–25 tok/s) is adequate but not ideal for interactive coding.
- Ingesting a prompt of 14K tokens takes ≈50 s, and each step ≈50 s — noticeable waits.
- 32K window; generation drops to 14–19 tok/s when it is full. KV cache is large (≈4.5 GB).
- Coding capability is below what interactive coding ideally needs — expect more mistakes or retries.
Why “Acceptable” rather than “Comfortable”?
- Workload suitability: Coding capability is below what interactive coding ideally needs — expect more mistakes or retries.
- Prompt processing: Ingesting the first prompt (14K tokens) takes ≈50 s; each of the ~4 follow-up turns takes ≈50 s (≈4 min per task).
- Generation: Expected 17–25 tok/s at ~16K tokens of context. Coding with repository context is comfortable from about 28 tok/s.
- Context: 32K window; generation drops to 14–19 tok/s when it is full. KV cache is large (≈4.5 GB).
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