Codegen 350M Mono — Hardware Requirements & GPU Compatibility
ChatCode
Codegen 350M Mono is a 350M-parameter open language model from Salesforce. It supports a context window of up to 2,048 tokens. At FP16 it needs about 0.77 GB of VRAM — see which GPUs and Macs can run it below.
Runs great on a 24 GB GPU
Runs from 8 GB (FP16 · 2K)Best on 24 GB (FP16 · ~851 tok/s · 2K)Macs from 8 GB (FP16 · ~93 tok/s)
BSD 3-Clause
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run Codegen 350M Mono
One recommended way. Every other method, OS and quant is in the step-by-step guide.Each way opens a step-by-step guide for your OS and quant.
Recommended
vllm serve Salesforce/codegen-350M-mono --max-model-len 2048vLLM repo with its own quantisation
Download: Weights on Hugging Face
Ways to runOther ways
each opens step-by-stepRun in cloud
Runs on RTX 3080 Ti (12 GB) (11 GB headroom) · FP16
- Generation speed
- ~770 tok/s
- generation speed
- Cost per 1M output tokens
- $0.044
- per 1M output tokens
How Much VRAM Does Codegen 350M Mono Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 0.8 GB | — | 16-bit · full — Full half-precision — baseline for inference | Fits: 2K context · ~851 tok/s |
Selected FP16
Full half-precision0.8GBat 2K context
- Weights
- 0.3 GB
- KV cache
- 0.2 GB
- Overhead
- 0.3 GB
Fits RTX 4090 (reference) · 23.2 GB free · ~851 tok/s
≥50 tok/s needs ≥59 GB/s · RTX 5090 ~1513, RTX 3090 Ti ~851
Which Hardware Can Run Codegen 350M Mono?
FP16 · 0.8 GBCodegen 350M Mono (FP16) requires 0.8 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 2+ GB is recommended. 52 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti. 163 devices with unified memory can run Codegen 350M Mono, including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB).
≤ 11 GB
≤ 11 GB tier →fits · 7.2 GB free · up to 2K ctx
16 GPUs · 1 Mac class · 2 device families
RTX 3080~642Simulate output on RTX 3080RTX 2080 Ti~520Simulate output on RTX 2080 TiRTX 3070 Ti~514Simulate output on RTX 3070 TiM3*8 GB~93Simulate output on iMac 24" M3 (8 GB)Orin Nano~86Simulate output on NVIDIA Jetson Orin Nano 8GB (Super)iPhone~62Simulate output on iPhone 1720 GB
fits · 19.2 GB free · up to 2K ctx
1 GPU
RX 7900 XT~623Simulate output on RX 7900 XT32–40 GB
fits · 31.2 GB free · up to 2K ctx
1 GPU · 3 Mac classes · 4 device families
RTX 5090~1513Simulate output on RTX 5090M5 Max36 GB~418Simulate output on Mac Studio M5 Max (36 GB)M532 GB~140Simulate output on MacBook Air 13" M5 (32 GB)Orin~173Simulate output on NVIDIA Jetson AGX Orin 32GBSnapdragon X Elite (X1E-80-100)~114Simulate output on Snapdragon X Elite Copilot+ PC
By VRAM tier · typical GPUs and Macs at FP16
Similar models
5| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| Granite 4.0 350MIBM | — | 1 GB | ~1150 | same size |
| SmolLM 360MHugging Face | — | 0.6 GB | ~1092 | same size |
| LFM2.5 350MLiquid AI | — | 0.8 GB | ~1130 | same size |
| ERNIE 4.5 0.3B PTBaidu | — | 0.7 GB | ~1213 | same size |
| LFM2 350MLiquid AI | — | 0.8 GB | ~1130 | same size |
Specifications
- Publisher
- Salesforce
- Parameters
- 350M
- Architecture
- CodeGenForCausalLM
- Context Length
- 2,048 tokens
- Vocabulary Size
- 51,200
- Release Date
- 2022-04-11
- License
- BSD 3-Clause
Frequently Asked Questions
How much VRAM does Codegen 350M Mono need?
Codegen 350M Mono requires 0.8 GB of VRAM at FP16.
VRAM = Weights + KV Cache + Overhead
Weights = 350M × 16 bits ÷ 8 = 0.7 GB
KV Cache + Overhead ≈ 0.1 GB (at 2K context + ~0.3 GB framework)
Fit ratings and hardware model lists check this model with room for a 16K-token context, which needs a little more memory.
VRAM usage by quantization
Can I run Codegen 350M Mono on a Mac?
Yes — MacBook Air 13" M3 (8 GB) and 145 other Macs can run Codegen 350M Mono. Apple Silicon uses unified memory, so the model shares RAM with the system. At FP16 you need at least 0.8 GB of usable unified memory (RAM minus macOS overhead).
Can I run Codegen 350M Mono locally?
Yes — Codegen 350M Mono can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is Codegen 350M Mono?
At FP16, Codegen 350M Mono can reach ~6234 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~851 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = (bandwidth GB/s ÷ model GB) × efficiency
Example: NVIDIA B200 → 8000 ÷ 0.77 × 0.65 = ~6753 tok/s
Estimated speed at FP16 (0.8 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of Codegen 350M Mono?
At FP16, the download is about 0.70 GB.
Which GPUs can run Codegen 350M Mono?
52 consumer GPUs can run Codegen 350M Mono at FP16. Top options include AMD Radeon RX 6700 XT, AMD Radeon RX 6800, AMD Radeon RX 6800 XT. 52 GPUs have plenty of headroom for comfortable inference.
Which devices can run Codegen 350M Mono?
163 devices with unified memory can run Codegen 350M Mono at FP16 (0.8 GB), including AMD Ryzen AI 9 HX 370 (Strix Point) Laptop, ASUS Ascent GX10, Beelink GTR9 Pro (Ryzen AI Max+ 395, 128 GB), Framework Desktop (Ryzen AI Max+ 395, 128 GB). Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.