Delphi 1e21 3.4Bparams 46.3Btokens Base Wc386k Lr1e5 Sft — Hardware Requirements & GPU Compatibility
Chat
Delphi 1e21 3.4Bparams 46.3Btokens Base Wc386k Lr1e5 Sft is a 142336-parameter open language model from laion in the Phi family. It supports a context window of up to 4,096 tokens. At Q4_K_M it needs about 0.85 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 · 4K)Best on 24 GB (Q4_K_M · ~771 tok/s · 4K)Macs from 8 GB (Q4_K_M · ~84 tok/s)
Other
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run Delphi 1e21 3.4Bparams 46.3Btokens Base Wc386k Lr1e5 Sft
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 laion/delphi-1e21-3.4Bparams-46.3Btokens-base-wc386k_lr1e5-sft --max-model-len 4096vLLM repo with its own quantisation
Download: Weights on Hugging Face
Ways to runOther ways
each opens step-by-stepRun in cloud
Runs on RTX 3060 12GB (10 GB headroom) · Q4_K_M
- Generation speed
- ~275 tok/s
- generation speed
- Cost per 1M output tokens
- $0.058
- per 1M output tokens
How Much VRAM Does Delphi 1e21 3.4Bparams 46.3Btokens Base Wc386k Lr1e5 Sft Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 0.8 GB | — | 2-bit · K-quant — 2-bit quantization with K-quant improvements | Fits: 4K context · ~771 tok/s | |
| 0.8 GB | — | 3-bit medium — 3-bit medium quantization | Fits: 4K context · ~771 tok/s | |
| 0.8 GB | — | 4-bit medium — 4-bit medium quantization — most popular sweet spot | Fits: 4K context · ~771 tok/s | |
| 0.8 GB | — | 5-bit medium — 5-bit medium quantization — good quality/size tradeoff | Fits: 4K context · ~771 tok/s | |
| 0.8 GB | — | 6-bit, very good — 6-bit quantization, very good quality | Fits: 4K context · ~771 tok/s | |
| 0.8 GB | — | 8-bit, near-lossless — 8-bit quantization, near-lossless | Fits: 4K context · ~771 tok/s | |
| 0.8 GB | — | 16-bit · bfloat — Brain floating point 16 — preferred for training | Fits: 4K context · ~771 tok/s |
Selected Q4_K_M
4-bit medium quantization1.4GBat 4K context
- Weights
- 0.1 GB
- KV cache
- 1.0 GB
- Overhead
- 0.3 GB
Fits RTX 4090 (reference) · 22.6 GB free · ~771 tok/s
≥50 tok/s needs ≥65 GB/s · RTX 5090 ~1370, RTX 3090 Ti ~771
Which Hardware Can Run Delphi 1e21 3.4Bparams 46.3Btokens Base Wc386k Lr1e5 Sft?
Q4_K_M · 0.8 GBDelphi 1e21 3.4Bparams 46.3Btokens Base Wc386k Lr1e5 Sft (Q4_K_M) 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. Using the full 4K context window can add up to 0.5 GB, bringing total usage to 1.4 GB. 52 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti. 163 devices with unified memory can run Delphi 1e21 3.4Bparams 46.3Btokens Base Wc386k Lr1e5 Sft, including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB).
≤ 11 GB
≤ 11 GB tier →fits · 6.6 GB free · up to 4K ctx
16 GPUs · 1 Mac class · 2 device families
RTX 3080~581Simulate output on RTX 3080RTX 2080 Ti~471Simulate output on RTX 2080 TiRTX 3070 Ti~465Simulate output on RTX 3070 TiM3*8 GB~84Simulate output on iMac 24" M3 (8 GB)Orin Nano~78Simulate output on NVIDIA Jetson Orin Nano 8GB (Super)iPhone~56Simulate output on iPhone 1720 GB
fits · 18.6 GB free · up to 4K ctx
1 GPU
RX 7900 XT~565Simulate output on RX 7900 XT32–40 GB
fits · 30.6 GB free · up to 4K ctx
1 GPU · 3 Mac classes · 4 device families
RTX 5090~1370Simulate output on RTX 5090M5 Max36 GB~379Simulate output on Mac Studio M5 Max (36 GB)M532 GB~127Simulate output on MacBook Air 13" M5 (32 GB)Orin~157Simulate output on NVIDIA Jetson AGX Orin 32GBSnapdragon X Elite (X1E-80-100)~103Simulate output on Snapdragon X Elite Copilot+ PC
By VRAM tier · typical GPUs and Macs at Q4_K_M
Alternatives to Delphi 1e21 3.4Bparams 46.3Btokens Base Wc386k Lr1e5 Sft
Phi family · 11 models →Specifications
- Publisher
- laion
- Family
- Phi
- Parameters
- 142336
- Architecture
- Qwen3ForCausalLM
- Context Length
- 4,096 tokens
- Vocabulary Size
- 128,256
- Release Date
- 2026-06-18
- License
- Other
Frequently Asked Questions
How much VRAM does Delphi 1e21 3.4Bparams 46.3Btokens Base Wc386k Lr1e5 Sft need?
Delphi 1e21 3.4Bparams 46.3Btokens Base Wc386k Lr1e5 Sft requires 0.8 GB of VRAM at Q4_K_M, or 0.8 GB at BF16. Full 4K context adds up to 0.5 GB (1.4 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 142336 × 4.8 bits ÷ 8 = 0 GB
KV Cache + Overhead ≈ 0.9 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 ≈ 1.4 GB (at full 4K context)
VRAM usage by quantization
What's the best quantization for Delphi 1e21 3.4Bparams 46.3Btokens Base Wc386k Lr1e5 Sft?
For Delphi 1e21 3.4Bparams 46.3Btokens Base Wc386k Lr1e5 Sft, Q4_K_M (0.8 GB) offers the best balance of quality and VRAM usage. Q5_K_M (0.8 GB) provides better quality if you have the VRAM. The smallest option is Q2_K at 0.8 GB.
VRAM requirement by quantization
★ Recommended — best balance of quality and VRAM usage.
Can I run Delphi 1e21 3.4Bparams 46.3Btokens Base Wc386k Lr1e5 Sft on a Mac?
Yes — MacBook Air 13" M3 (8 GB) and 145 other Macs can run Delphi 1e21 3.4Bparams 46.3Btokens Base Wc386k Lr1e5 Sft. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 0.8 GB of usable unified memory (RAM minus macOS overhead).
Can I run Delphi 1e21 3.4Bparams 46.3Btokens Base Wc386k Lr1e5 Sft locally?
Yes — Delphi 1e21 3.4Bparams 46.3Btokens Base Wc386k Lr1e5 Sft can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is Delphi 1e21 3.4Bparams 46.3Btokens Base Wc386k Lr1e5 Sft?
At Q4_K_M, Delphi 1e21 3.4Bparams 46.3Btokens Base Wc386k Lr1e5 Sft can reach ~5647 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~771 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 ÷ 0.85 × 0.65 = ~6118 tok/s
Estimated speed at Q4_K_M (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 Delphi 1e21 3.4Bparams 46.3Btokens Base Wc386k Lr1e5 Sft?
At Q4_K_M, the download is about 0.01 GB. The full-precision BF16 version is 0.01 GB. The smallest option (Q2_K) is 0.01 GB.
Which GPUs can run Delphi 1e21 3.4Bparams 46.3Btokens Base Wc386k Lr1e5 Sft?
52 consumer GPUs can run Delphi 1e21 3.4Bparams 46.3Btokens Base Wc386k Lr1e5 Sft at Q4_K_M. 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 Delphi 1e21 3.4Bparams 46.3Btokens Base Wc386k Lr1e5 Sft?
163 devices with unified memory can run Delphi 1e21 3.4Bparams 46.3Btokens Base Wc386k Lr1e5 Sft at Q4_K_M (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.