Internlm 7B — Hardware Requirements & GPU Compatibility
Chat
InternLM-7B is InternLM's open base pretrained language model, not instruction-tuned, built at 7 billion parameters and trained on trillions of high-quality tokens to serve as a general-purpose knowledge foundation for downstream fine-tuning. On the OpenCompass evaluation suite it outperformed same-size peers such as LLaMA-7B and Baichuan-7B across disciplinary, language, knowledge, reasoning, and comprehension benchmarks. A matching InternLM-Chat-7B instruction-tuned version was released alongside it. Its 7B size fits on a single consumer GPU. Context length is 2,048 tokens. The code is released under Apache 2.0, while the model weights are free for academic research and free for commercial use only after completing InternLM's application form. It was published in July 2023, as InternLM's first open base model; larger InternLM2 and later families followed.
2.5K downloads
96 likesSpecifications
- Publisher
- InternLM
- Family
- InternLM
- Parameters
- 7B
- Architecture
- InternLMForCausalLM
- Context Length
- 2,048 tokens
- Vocabulary Size
- 103,168
- Release Date
- 2023-07-06
Get Started
HuggingFace
How Much VRAM Does Internlm 7B Need?
Select a quantization to see compatible GPUs below.
| Quantization | Bits | VRAM | + Context | File Size | Quality |
|---|---|---|---|---|---|
| FP16est. | 16.00 | 15.4 GB | — | 14.00 GB | Full half-precision — baseline for inference |
est.= calculated VRAM estimate; no published GGUF file found for that quantization yet. Other rows are verified against real community uploads.
Which GPUs Can Run Internlm 7B?
FP16 · 15.4 GBInternlm 7B (FP16) requires 15.4 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 21+ GB is recommended. 26 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti, NVIDIA GeForce RTX 5080.
Runs great
— Plenty of headroom
Decent
— Enough VRAM, may be tightWhich Devices Can Run Internlm 7B?
FP16 · 15.4 GB47 devices with unified memory can run Internlm 7B, including NVIDIA DGX H100, NVIDIA DGX A100 640GB, Mac Mini M4 (16 GB).
Runs great
— Plenty of headroomBenchmarks
Benchmark details →Related Models
Frequently Asked Questions
- How much VRAM does Internlm 7B need?
Internlm 7B requires 15.4 GB of VRAM at FP16.
VRAM = Weights + KV Cache + Overhead
Weights = 7B × 16 bits ÷ 8 = 14 GB
KV Cache + Overhead ≈ 1.4 GB (at 2K context + ~0.3 GB framework)
VRAM usage by quantization
FP1615.4 GB- Can I run Internlm 7B on a Mac?
Internlm 7B requires at least 15.4 GB at FP16, which exceeds the unified memory of most consumer Macs. You would need a Mac Studio or Mac Pro with a high-memory configuration.
- Can I run Internlm 7B locally?
Yes — Internlm 7B can run locally on consumer hardware. At FP16 quantization it needs 15.4 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.
- How fast is Internlm 7B?
At FP16, Internlm 7B can reach ~312 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~43 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 ÷ 15.4 × 0.65 = ~338 tok/s
Estimated speed at FP16 (15.4 GB)
~338 tok/s~43 tok/s~338 tok/s~312 tok/sReal-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
- What's the download size of Internlm 7B?
At FP16, the download is about 14.00 GB.
- Which GPUs can run Internlm 7B?
26 consumer GPUs can run Internlm 7B at FP16 (15.4 GB). Top options include AMD Radeon RX 7900 XTX, NVIDIA GeForce RTX 3090, NVIDIA GeForce RTX 3090 Ti, AMD Radeon RX 6800. 7 GPUs have plenty of headroom for comfortable inference.
- Which devices can run Internlm 7B?
49 devices with unified memory can run Internlm 7B at FP16 (15.4 GB), including AMD Ryzen AI 9 HX 370 (Strix Point) Laptop, ASUS Ascent GX10, Asus ROG Flow Z13 (2025, Ryzen AI Max+ 395, 128 GB), Beelink GTR9 Pro (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.