CodeGeeX 4 9B needs ~9.6 GB VRAM. MacBook Pro M3 24GB has 17.3 GB. With Q4_K_M quantization, expect ~12 tok/s.
Operating mode
Interactive favors responsiveness, while light API and scale-out lean harder on serving readiness. The fit stays the same, but the recommendation lens changes.
Current mode
Balanced
Balanced for general local use. Keeps the ranking neutral across personal and serving workflows.
Select quantization to explore
Fit status
Runs well
Decode
13.5 tok/s
TTFT
14291 ms
Safe context
131K
Memory
9.6 GB / 17.3 GB
This setup is broadly balanced for this model.
Shared-memory contention still exists
The OS, browser, and inference runtime all compete for the same physical memory pool, so real-world headroom is less forgiving than raw capacity suggests.
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Chat | A | Runs well | 12.4 tok/s | 8526 ms | 131K |
| Coding | A | Runs well | 12.4 tok/s | 15630 ms | 131K |
| Agentic Coding | A | Runs well | 12.4 tok/s | 22735 ms | 131K |
| Reasoning | A | Runs well | 12.4 tok/s | 18472 ms | 131K |
| RAG | A | Runs well | 12.4 tok/s | 28419 ms | 131K |
How CodeGeeX 4 9B (9B params) fits at each quantization level on MacBook Pro M3 24GB (17.3 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 3.5 GB | Low | A75 |
Q3_K_S | 3 | 4.4 GB | Low | A75 |
NVFP4 | 4 |
Copy-paste commands to run CodeGeeX 4 9B on your machine.
Run
docker run --rm -it ghcr.io/ggerganov/llama.cpp:full \
--hf-repo "THUDM/codegeex4-all-9b" \
--hf-file "codegeex4-all-9b-Q4_K_M.gguf" \
-c 4096 -ngl 99Your hardware
| Model | Params | Grade | Decode | Capabilities |
|---|---|---|---|---|
| 24B | B | 3.8 tok/s | ||
| 24B | B | 3.8 tok/s |
Yes, MacBook Pro M3 24GB can run CodeGeeX 4 9B with a A grade (Runs well). Expected decode speed: 12.4 tok/s.
CodeGeeX 4 9B (9B parameters) requires approximately 9.6 GB of memory with Q4_K_M quantization.
The recommended quantization for CodeGeeX 4 9B is Q4_K_M, which balances quality and memory efficiency.
On MacBook Pro M3 24GB, CodeGeeX 4 9B achieves approximately 12.4 tokens per second decode speed with a time-to-first-token of 15630ms using Q4_K_M quantization.
For coding workloads, CodeGeeX 4 9B on MacBook Pro M3 24GB receives a A grade with 12.4 tok/s and 131K context.
On MacBook Pro M3 24GB, CodeGeeX 4 9B can safely use up to 131K tokens of context. The model's official context limit is 131K, but available memory constrains the safe maximum.
Paste this snippet into any page to show a live fit card.
<iframe src="https://willitrunai.com/embed/codegeex-4-9b-on-m3-24gb" width="400" height="180" frameborder="0" style="border:none;border-radius:12px;overflow:hidden;" title="Will It Run AI — fit result"></iframe>
Preview:
5.0 GB |
| Medium |
| A76 |
Q4_K_M | 4 | 5.5 GB | Medium | A76 |
Q5_K_M | 5 | 6.5 GB | High | A77 |
Q6_K | 6 | 7.4 GB | High | A78 |
Q8_0Best for your GPU | 8 | 9.6 GB | Very High | A79 |
F16 | 16 | 18.5 GB | Maximum | F0 |
| 14B | S | 8.6 tok/s |
| 14.7B | S | 8.2 tok/s |
| 24B | B | 3.8 tok/s |
Not always. MacBook Pro M3 24GB can often fit larger models thanks to unified memory, but a discrete GPU with dedicated high-bandwidth VRAM may still decode faster once the model fits. For this combination, the important distinction is capacity versus sustained throughput.