Qwen2.5 Coder 14B — Hardware Requirements & GPU Compatibility
ChatCode
Qwen2.5 Coder 14B is a 14.8B-parameter open language model from Alibaba in the Qwen 2.5 family. It supports a context window of up to 32,768 tokens. At Q4_K_M it needs about 9.56 GB of VRAM — see which GPUs and Macs can run it below.
Runs great on a 24 GB GPU
Runs from 8 GB (IQ2_XS · 16K)Best on 24 GB (Q4_K_M · ~69 tok/s · 32K)Macs from 16 GB (Q4_K_M · ~15 tok/s)
Apache 2.0
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run Qwen2.5 Coder 14B
One recommended way. Every other method, OS and quant is in the step-by-step guide.
Recommended
ollama run hf.co/bartowski/Qwen2.5-Coder-14B-GGUF:Q4_K_MQ4_K_M fits fully on the RTX 4090 · 11.6 GB free · 16K context
Download: Qwen2.5-Coder-14B-Q4_K_M.gguf8.99 GBbartowski
Other ways
each opens step-by-stepRun in cloud
Fits on RTX 5060 Ti 16GB (3 GB headroom) · Q4_K_M
- Generation speed
- ~31 tok/s
- generation speed
- Cost per 1M output tokens
- $1.50
- per 1M output tokens
How Much VRAM Does Qwen2.5 Coder 14B Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 7.0 GB | 13 GB | 2-bit · K-quant — 2-bit quantization with K-quant improvements | Fits: 32K context · ~94 tok/s | |
| 7.2 GB | 13.2 GB | 3-bit small — 3-bit small quantization | Fits: 32K context · ~92 tok/s | |
| 7.9 GB | 13.9 GB | 3-bit medium — 3-bit medium quantization | Fits: 32K context · ~83 tok/s | |
| 8.1 GB | 14.1 GB | 4-bit · legacy — 4-bit legacy quantization | Fits: 32K context · ~81 tok/s | |
| 9.6 GB | 15.6 GB | 4-bit medium — 4-bit medium quantization — most popular sweet spot | Fits: 32K context · ~69 tok/s | |
| 11.2 GB | 17.3 GB | 5-bit medium — 5-bit medium quantization — good quality/size tradeoff | Fits: 32K context · ~58 tok/s | |
| 12.9 GB | 18.9 GB | 6-bit, very good — 6-bit quantization, very good quality | Fits: 32K context · ~51 tok/s | |
| 15.5 GB | 21.5 GB | 8-bit, near-lossless — 8-bit quantization, near-lossless | Fits: 32K context · ~42 tok/s |
Selected Q4_K_M
4-bit medium quantization12.4GBat 16K context
- Weights
- 8.9 GB
- KV cache
- 3.2 GB
- Overhead
- 0.3 GB
Fits RTX 4090 (reference) · 11.6 GB free · ~69 tok/s
≥50 tok/s needs ≥735 GB/s · RTX 5090 ~122, RTX 3090 Ti ~69
Which Hardware Can Run Qwen2.5 Coder 14B?
Q4_K_M · 9.6 GBQwen2.5 Coder 14B (Q4_K_M) requires 9.6 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 13+ GB is recommended. Using the full 33K context window can add up to 6.0 GB, bringing total usage to 15.6 GB. 40 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti, NVIDIA GeForce RTX 3080 Ti. 150 devices with unified memory can run Qwen2.5 Coder 14B, including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB), Mac Mini M2 Pro (16 GB).
16–18 GB
16–18 GB tier →fits · 3.6 GB free · up to 32K ctx
18 GPUs · 2 Mac classes · 2 device families
RTX 5080~65Simulate output on RTX 5080RTX 5070 Ti~61Simulate output on RTX 5070 TiRTX 4080 SUPER~50Simulate output on RTX 4080 SUPERM516 GB~11Simulate output on MacBook Air 13" M5 (16 GB)M3 Pro*18 GB~11Simulate output on MacBook Pro 14" M3 Pro (18 GB)iPad~11Simulate output on iPad Pro M5 13" (16 GB)Orin NX~7Simulate output on NVIDIA Jetson Orin NX 16GB20 GB
fits · 7.6 GB free · up to 32K ctx
1 GPU
RX 7900 XT~50Simulate output on RX 7900 XT32–40 GB
fits · 19.6 GB free · up to 32K ctx
1 GPU · 3 Mac classes · 4 device families
RTX 5090~122Simulate output on RTX 5090M5 Max36 GB~34Simulate output on Mac Studio M5 Max (36 GB)M532 GB~11Simulate output on MacBook Air 13" M5 (32 GB)Orin~14Simulate output on NVIDIA Jetson AGX Orin 32GBSnapdragon X Elite (X1E-80-100)~9Simulate output on Snapdragon X Elite Copilot+ PC
By VRAM tier · typical GPUs and Macs at Q4_K_M
How good is Qwen2.5 Coder 14B?
llmrun Score blends public benchmark boards into one number; 100 = the average open model.
Not enough data yet for an llmrun Score.
A Score needs results on enough independent boards. How the Composite Score is built
Benchmarks
Here are all 5 boards that have results for this model, sorted by their weight in the llmrun Score; badges show who ran the test.
- MMLU
Epoch AI · Sep 2024Lab-reported
Lab-reported75.2%
#21/85 open
- GSM8K
Epoch AI · Sep 2024Lab-reported
Lab-reported88.7%
#6/62 open
- HellaSwag
Epoch AI · Sep 2024Lab-reported
Lab-reported80.2%
#16/47 open
- ARC Challenge
Epoch AI · Sep 2024Lab-reported
Lab-reported66.0%
#21/56 open
- WinoGrande
Epoch AI · Sep 2024Lab-reported
Lab-reported76.8%
#16/51 open
Alternatives to Qwen2.5 Coder 14B
Qwen 2.5 family · 43 models →Similar models
6| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| Apriel 1.6 15B ThinkerServiceNow-AI | — | 13.2 GB | ~67 | same size |
| Phi 4 ReasoningMicrosoft | — | 12.5 GB | ~69 | same size |
| Phi 3 Medium 128k InstructMicrosoft | — | 12 GB | ~72 | same size |
| JiRackUltra 14BCMSManhattan | — | 12.4 GB | ~69 | same size |
| Nemotron Terminal 14BNVIDIA | — | 11.9 GB | ~69 | same size |
| Ministral 3 14B Reasoning 2512Mistral AI | — | 12 GB | ~72 | same size |
Fine-tunes & variants(1)
Specifications
- Publisher
- Alibaba
- Family
- Qwen 2.5
- Parameters
- 14.8B
- Architecture
- Qwen2ForCausalLM
- Context Length
- 32,768 tokens
- Vocabulary Size
- 152,064
- Release Date
- 2024-11-08
- License
- Apache 2.0
Frequently Asked Questions
How much VRAM does Qwen2.5 Coder 14B need?
Qwen2.5 Coder 14B requires 9.6 GB of VRAM at Q4_K_M, or 30.2 GB at BF16. Full 33K context adds up to 6.0 GB (15.6 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 14.8B × 4.8 bits ÷ 8 = 8.9 GB
KV Cache + Overhead ≈ 0.7 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.
KV Cache + Overhead ≈ 6.7 GB (at full 33K context)
VRAM usage by quantization
Can NVIDIA GeForce RTX 4090 run Qwen2.5 Coder 14B?
Yes, at Q8_0 (15.5 GB) or lower. Higher quantizations like BF16 (30.2 GB) exceed the NVIDIA GeForce RTX 4090's 24 GB.
What's the best quantization for Qwen2.5 Coder 14B?
For Qwen2.5 Coder 14B, Q4_K_M (9.6 GB) offers the best balance of quality and VRAM usage. Q4_K_L (9.8 GB) provides better quality if you have the VRAM. The smallest option is IQ2_XS at 5.1 GB.
VRAM requirement by quantization
★ Recommended — best balance of quality and VRAM usage.
Can I run Qwen2.5 Coder 14B on a Mac?
Yes — Mac Mini M2 Pro (16 GB) and 133 other Macs can run Qwen2.5 Coder 14B. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 9.6 GB of usable unified memory (RAM minus macOS overhead).
Can I run Qwen2.5 Coder 14B locally?
Yes — Qwen2.5 Coder 14B can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is Qwen2.5 Coder 14B?
At Q4_K_M, Qwen2.5 Coder 14B can reach ~502 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~69 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 ÷ 9.56 × 0.65 = ~544 tok/s
Estimated speed at Q4_K_M (9.6 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of Qwen2.5 Coder 14B?
At Q4_K_M, the download is about 8.86 GB. The full-precision BF16 version is 29.54 GB. The smallest option (IQ2_XS) is 4.43 GB.
Which GPUs can run Qwen2.5 Coder 14B?
40 consumer GPUs can run Qwen2.5 Coder 14B at Q4_K_M. Top options include AMD Radeon RX 6800, AMD Radeon RX 6800 XT, AMD Radeon RX 6900 XT, AMD Radeon RX 6700 XT. 26 GPUs have plenty of headroom for comfortable inference.
Which devices can run Qwen2.5 Coder 14B?
150 devices with unified memory can run Qwen2.5 Coder 14B at Q4_K_M (9.6 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.