TwIL 8B — Hardware Requirements & GPU Compatibility
ChatReasoningMathCodeFunctions
TwIL 8B is a 8.8B-parameter open language model from webAI-Official. It supports a context window of up to 131,072 tokens. At Q4_K_M it needs about 5.91 GB of VRAM — see which GPUs and Macs can run it below.
Runs great on a 24 GB GPU
Runs from 8 GB (Q4_K_M · 8K)Best on 24 GB (Q4_K_M · ~111 tok/s · 64K)Macs from 16 GB (Q4_K_M · ~24 tok/s)
Other
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run TwIL 8B
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 webAI-Official/TwIL-8B --max-model-len 16384vLLM 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 (10 GB) (1 GB headroom) · Q4_K_M
- Generation speed
- ~84 tok/s
- generation speed
- Cost per 1M output tokens
- $0.37
- per 1M output tokens
How Much VRAM Does TwIL 8B Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 4.4 GB | 9.4 GB | 2-bit · K-quant — 2-bit quantization with K-quant improvements | Fits: 64K context · ~150 tok/s | |
| 4.9 GB | 10 GB | 3-bit medium — 3-bit medium quantization | Fits: 64K context · ~133 tok/s | |
| 5.9 GB | 10.9 GB | 4-bit medium — 4-bit medium quantization — most popular sweet spot | Fits: 64K context · ~111 tok/s | |
| 6.9 GB | 11.9 GB | 5-bit medium — 5-bit medium quantization — good quality/size tradeoff | Fits: 64K context · ~95 tok/s | |
| 7.9 GB | 12.9 GB | 6-bit, very good — 6-bit quantization, very good quality | Fits: 64K context · ~83 tok/s | |
| 9.4 GB | 14.5 GB | 8-bit, near-lossless — 8-bit quantization, near-lossless | Fits: 64K context · ~70 tok/s | |
| 18.2 GB | 23.3 GB | 16-bit · bfloat — Brain floating point 16 — preferred for training | Fits: 32K context · ~36 tok/s |
Selected Q4_K_M
4-bit medium quantization8.3GBat 16K context
- Weights
- 5.3 GB
- KV cache
- 2.7 GB
- Overhead
- 0.3 GB
Fits RTX 4090 (reference) · 15.7 GB free · ~111 tok/s
≥50 tok/s needs ≥454 GB/s · RTX 5090 ~197, RTX 3090 Ti ~111
Which Hardware Can Run TwIL 8B?
Q4_K_M · 5.9 GBTwIL 8B (Q4_K_M) requires 5.9 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 8+ GB is recommended. Using the full 131K context window can add up to 21.1 GB, bringing total usage to 27.1 GB. 52 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti, NVIDIA GeForce RTX 3070 Ti. 151 devices with unified memory can run TwIL 8B, including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB), NVIDIA Jetson Orin Nano 8GB (Super).
12 GB
12 GB tier →fits · 3.7 GB free · up to 32K ctx
10 GPUs · 1 device family
RTX 3080 Ti~100Simulate output on RTX 3080 TiRTX 5070~74Simulate output on RTX 5070RTX 4070~55Simulate output on RTX 4070iPhone~9Simulate output on Apple iPhone 17 Pro20 GB
fits · 11.7 GB free · up to 64K ctx
1 GPU
RX 7900 XT~81Simulate output on RX 7900 XT32–40 GB
fits · 23.7 GB free · up to 128K ctx
1 GPU · 3 Mac classes · 4 device families
RTX 5090~197Simulate output on RTX 5090M5 Max36 GB~55Simulate output on Mac Studio M5 Max (36 GB)M532 GB~18Simulate output on MacBook Air 13" M5 (32 GB)Orin~23Simulate output on NVIDIA Jetson AGX Orin 32GBSnapdragon X Elite (X1E-80-100)~15Simulate output on Snapdragon X Elite Copilot+ PC
By VRAM tier · typical GPUs and Macs at Q4_K_M
Similar models
6| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| Granite Switch 4.1 8BIBM | — | 8.4 GB | ~111 | same size |
| Apertus V1.5 8Bswiss-ai | — | 8.5 GB | ~111 | same size |
| Ornith 1.0 9BDeep Reinforce | — | 6.5 GB | ~106 | same size |
| LFM2.5 8B A1BLiquid AI | — | 5.6 GB | ~335 | same size |
| MiniCPM O 4 5OpenBMB | — | 8.3 GB | ~105 | same size |
| Qwen3 VL 8B InstructAlibaba | — | 8 GB | ~112 | same size |
Specifications
- Publisher
- webAI-Official
- Parameters
- 8.8B
- Architecture
- GraniteForCausalLM
- Context Length
- 131,072 tokens
- Vocabulary Size
- 100,352
- Release Date
- 2026-10-14
- License
- Other
Frequently Asked Questions
How much VRAM does TwIL 8B need?
TwIL 8B requires 5.9 GB of VRAM at Q4_K_M, or 18.2 GB at BF16. Full 131K context adds up to 21.1 GB (27.1 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 8.8B × 4.8 bits ÷ 8 = 5.3 GB
KV Cache + Overhead ≈ 0.6 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 ≈ 21.8 GB (at full 131K context)
VRAM usage by quantization
What's the best quantization for TwIL 8B?
For TwIL 8B, Q4_K_M (5.9 GB) offers the best balance of quality and VRAM usage. Q5_K_M (6.9 GB) provides better quality if you have the VRAM. The smallest option is Q2_K at 4.4 GB.
VRAM requirement by quantization
★ Recommended — best balance of quality and VRAM usage.
Can I run TwIL 8B on a Mac?
Yes — Mac Mini M2 Pro (16 GB) and 133 other Macs can run TwIL 8B. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 5.9 GB of usable unified memory (RAM minus macOS overhead).
Can I run TwIL 8B locally?
Yes — TwIL 8B can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is TwIL 8B?
At Q4_K_M, TwIL 8B can reach ~812 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~111 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 ÷ 5.91 × 0.65 = ~880 tok/s
Estimated speed at Q4_K_M (5.9 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of TwIL 8B?
At Q4_K_M, the download is about 5.27 GB. The full-precision BF16 version is 17.58 GB. The smallest option (Q2_K) is 3.74 GB.
Which GPUs can run TwIL 8B?
52 consumer GPUs can run TwIL 8B at Q4_K_M. Top options include AMD Radeon RX 6700 XT, AMD Radeon RX 6800, AMD Radeon RX 6800 XT, AMD Radeon RX 7600. 40 GPUs have plenty of headroom for comfortable inference.
Which devices can run TwIL 8B?
151 devices with unified memory can run TwIL 8B at Q4_K_M (5.9 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.