Baichuan M3 235B needs ~132.9 GB VRAM. AMD Instinct MI250 128GB has 128.0 GB. With Q2_K quantization, expect ~14 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
56.6 GB over capacity — needs offload or smaller quantization
Fit status
Too heavy
Decode
5.3 tok/s
TTFT
36751 ms
Safe context
4K
Memory
184.6 GB / 128.0 GB
Offload
30%
This setup is broadly balanced for this model.
Very little memory headroom
You can run the model, but there is not much room left for longer context, bigger batches, extra apps, or future model updates.
Buy headroom, not only minimum fit
A slightly larger memory tier gives you safer context growth and makes the recommendation more future-proof.
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Chat | F | Too heavy | 6.2 tok/s | 17028 ms | 4K |
| Coding | F | Too heavy | 5.3 tok/s | 36751 ms | 4K |
| Agentic Coding | F | Too heavy | 3.9 tok/s | 71635 ms | 4K |
| Reasoning | F | Too heavy | 5.3 tok/s | 43433 ms | 4K |
| RAG | F | Too heavy | 3.9 tok/s | 89543 ms | 4K |
Inference speed
Estimated decode speed (tokens/sec) for Baichuan M3 235B at Q4_K_M across popular GPUs and Apple Silicon, including multi-GPU rigs, using the fastest local runtime per device. Fastest is Mac Studio M3 Ultra 256GB at ~3 tok/s. Speed is memory-bandwidth bound, so cards that fit the whole model in VRAM run far faster than ones that offload to system RAM.
| GPU / Mac | Memory | Quant | Speed (tok/s) | Fits? |
|---|---|---|---|---|
Mac Studio M3 Ultra 256GB | 256 GB | Q4_K_M | 3.4 | Heavy offload |
 RTX 5090 32GB | 32 GB | Q4_K_M | 2.0 | Too big |
| RTX 4090 24GB | 24 GB | Q4_K_M | 2.0 | Too big |
| RTX 4080 Super 16GB | 16 GB | Q4_K_M | 2.0 | Too big |
| RTX 3090 24GB | 24 GB | Q4_K_M | 2.0 | Too big |
| RTX 4070 12GB | 12 GB | Q4_K_M | 2.0 | Too big |
| RTX 3060 12GB | 12 GB | Q4_K_M | 2.0 | Too big |
| RTX 4060 8GB | 8 GB | Q4_K_M | 2.0 | Too big |
RX 7900 XTX 24GB | 24 GB | Q4_K_M | 2.0 | Too big |
MacBook Pro M4 Max 128GB | 128 GB | Q4_K_M | 2.0 | Too big |
Mac Studio M2 Ultra 128GB | 128 GB | Q4_K_M | 2.0 | Too big |
Mac Studio M1 Ultra 128GB | 128 GB | Q4_K_M | 2.0 | Too big |
MacBook Pro M4 Max 64GB | 64 GB | Q4_K_M | 2.0 | Too big |
MacBook Pro M3 Max 64GB | 64 GB | Q4_K_M | 2.0 | Too big |
MacBook Pro M1 Max 64GB | 64 GB | Q4_K_M | 2.0 | Too big |
MacBook Pro M4 Pro 48GB | 48 GB | Q4_K_M | 2.0 | Too big |
| 2× RTX 4090 24GB | 48 GB | Q4_K_M | 2.0 | Too big |
| 2× RTX 3090 24GB | 48 GB | Q4_K_M | 2.0 | Too big |
2× RX 7900 XTX 24GB | 48 GB | Q4_K_M | 2.0 | Too big |
| 4× RTX 3060 12GB | 48 GB | Q4_K_M | 2.0 | Too big |
Estimates for single-stream decoding at Q4_K_M; real tokens/sec varies with prompt length, context, batch size, and runtime build. Prompt processing (prefill) is faster than the decode figures shown here.
How Baichuan M3 235B (235B params) fits at each quantization level on AMD Instinct MI250 128GB (128.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_KBest for your GPU | 2 | 91.7 GB | Low | C47 |
Q3_K_S | 3 | 115.2 GB | Low | F0 |
NVFP4 | 4 | 131.6 GB | Medium | F0 |
Q4_K_M | 4 | 143.4 GB | Medium | F0 |
Q5_K_M | 5 | 169.2 GB | High | F0 |
Q6_K | 6 | 192.7 GB | High | F0 |
Q8_0 | 8 | 251.5 GB | Very High | F0 |
F16 | 16 | 481.7 GB | Maximum | F0 |
Copy-paste commands to run Baichuan M3 235B on your machine.
Run
lms load hf-mradermacher--baichuan-m3-235b-gguf && lms server startUpgrade options
Makes the model fit on the accelerator instead of staying completely out of reach.
Removes host-memory offload, which is usually the single biggest latency and throughput win.
~$8,000 MSRP
Makes the model fit on the accelerator instead of staying completely out of reach.
Removes host-memory offload, which is usually the single biggest latency and throughput win.
~$15,000 MSRP
Makes the model fit on the accelerator instead of staying completely out of reach.
Removes host-memory offload, which is usually the single biggest latency and throughput win.
~$20,000 MSRP
Yes, AMD Instinct MI250 128GB can run Baichuan M3 235B at Q2_K quantization (Runs with offload (needs ~3.4 GB host RAM)). The recommended Q4_K_M requires 184.6 GB which exceeds available memory, but at Q2_K it needs only 132.9 GB. Expected decode speed: 14.0 tok/s.
Baichuan M3 235B (235B parameters) requires approximately 184.6 GB at Q4_K_M quantization. On AMD Instinct MI250 128GB, it fits at Q2_K using 132.9 GB.
The recommended quantization is Q4_K_M, but on AMD Instinct MI250 128GB the best fitting quantization is Q2_K, which uses 132.9 GB.
On AMD Instinct MI250 128GB, Baichuan M3 235B achieves approximately 14.0 tokens per second decode speed with a time-to-first-token of 13841ms using Q2_K quantization.
For coding workloads, Baichuan M3 235B on AMD Instinct MI250 128GB receives a F grade with 5.3 tok/s and 4K context.
On AMD Instinct MI250 128GB, Baichuan M3 235B can safely use up to 13K tokens of context at Q2_K quantization. The model's official context limit is —, but available memory constrains the safe maximum.
Buy headroom, not only minimum fit. A slightly larger memory tier gives you safer context growth and makes the recommendation more future-proof.
Paste this snippet into any page to show a live fit card.
<iframe src="https://willitrunai.com/embed/hf-mradermacher--baichuan-m3-235b-gguf-on-instinct-mi250-128gb" width="400" height="180" frameborder="0" style="border:none;border-radius:12px;overflow:hidden;" title="Will It Run AI — fit result"></iframe>
Preview: