SOLAR 10.7B Instruct v1.0 — Hardware Requirements & GPU Compatibility
Chat
SOLAR 10.7B Instruct v1.0 is a 10.7B-parameter open language model from Upstage in the Solar family. It supports a context window of up to 4,096 tokens. At Q4_K_M it needs about 7.14 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 · ~92 tok/s · 4K)Macs from 16 GB (Q4_K_M · ~20 tok/s)
CC BY-NC 4.0
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run SOLAR 10.7B Instruct v1.0
One recommended way. Every other method, OS and quant is in the step-by-step guide.
Recommended
vllm serve upstage/SOLAR-10.7B-Instruct-v1.0 --max-model-len 4096vLLM repo with its own quantisation
Download: Weights on Hugging Face
Other ways
each opens step-by-stepRun in cloud
Fits on RTX 3060 12GB (4 GB headroom) · Q4_K_M
- Generation speed
- ~33 tok/s
- generation speed
- Cost per 1M output tokens
- $0.70
- per 1M output tokens
How Much VRAM Does SOLAR 10.7B Instruct v1.0 Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 5.3 GB | — | 2-bit · K-quant — 2-bit quantization with K-quant improvements | Fits: 4K context · ~125 tok/s | |
| 5.9 GB | — | 3-bit medium — 3-bit medium quantization | Fits: 4K context · ~111 tok/s | |
| 7.1 GB | — | 4-bit medium — 4-bit medium quantization — most popular sweet spot | Fits: 4K context · ~92 tok/s | |
| 8.3 GB | — | 5-bit medium — 5-bit medium quantization — good quality/size tradeoff | Fits: 4K context · ~79 tok/s | |
| 9.6 GB | — | 6-bit, very good — 6-bit quantization, very good quality | Fits: 4K context · ~69 tok/s | |
| 11.4 GB | — | 8-bit, near-lossless — 8-bit quantization, near-lossless | Fits: 4K context · ~57 tok/s | |
| 22.2 GB | — | 16-bit · full — Full half-precision — baseline for inference | Fits: 4K context · ~30 tok/s |
Selected Q4_K_M
4-bit medium quantization7.5GBat 4K context
- Weights
- 6.4 GB
- KV cache
- 0.8 GB
- Overhead
- 0.3 GB
Fits RTX 4090 (reference) · 16.5 GB free · ~92 tok/s
≥50 tok/s needs ≥549 GB/s · RTX 5090 ~163, RTX 3090 Ti ~92
Which Hardware Can Run SOLAR 10.7B Instruct v1.0?
Q4_K_M · 7.1 GBSOLAR 10.7B Instruct v1.0 (Q4_K_M) requires 7.1 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 10+ GB is recommended. Using the full 4K context window can add up to 0.4 GB, bringing total usage to 7.5 GB. 52 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti, NVIDIA GeForce RTX 3080. 151 devices with unified memory can run SOLAR 10.7B Instruct v1.0, including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB), NVIDIA Jetson Orin Nano 8GB (Super).
≤ 11 GB
≤ 11 GB tier →fits · 0.5 GB free · up to 4K ctx
16 GPUs · 1 device family
RTX 3080~69Simulate output on RTX 3080RTX 2080 Ti~56Simulate output on RTX 2080 TiRTX 3070 Ti~55Simulate output on RTX 3070 TiOrin Nano~9Simulate output on NVIDIA Jetson Orin Nano 8GB (Super)20 GB
fits · 12.5 GB free · up to 4K ctx
1 GPU
RX 7900 XT~67Simulate output on RX 7900 XT32–40 GB
fits · 24.5 GB free · up to 4K ctx
1 GPU · 3 Mac classes · 4 device families
RTX 5090~163Simulate output on RTX 5090M5 Max36 GB~45Simulate output on Mac Studio M5 Max (36 GB)M532 GB~15Simulate output on MacBook Air 13" M5 (32 GB)Orin~19Simulate output on NVIDIA Jetson AGX Orin 32GBSnapdragon X Elite (X1E-80-100)~12Simulate output on Snapdragon X Elite Copilot+ PC
By VRAM tier · typical GPUs and Macs at Q4_K_M
How good is SOLAR 10.7B Instruct v1.0?
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
This is the single board showing results for this model; badges show who ran the test.
- LMArena Text
LMArena · Oct 2026Arena
Arena1152.5
#167/187 open
Similar models
6| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| GLM 4.1V 9B ThinkingZ.ai | — | 7.2 GB | ~100 | same size |
| Qwen2.5 Omni 7BAlibaba | — | 9.9 GB | ~93 | same size |
| INTELLECT 1 InstructPrimeIntellect | — | 7.8 GB | ~97 | same size |
| Llama 3.2 11B Vision InstructMeta | — | 9.8 GB | ~93 | same size |
| Falcon 11BTII UAE | — | 8.5 GB | ~89 | same size |
| GigaChat3.1 Audio 10B A1.8Bai-sage | — | 22.9 GB | ~198 | same size |
Specifications
- Publisher
- Upstage
- Family
- Solar
- Parameters
- 10.7B
- Architecture
- LlamaForCausalLM
- Context Length
- 4,096 tokens
- Vocabulary Size
- 32,000
- Release Date
- 2023-12-12
- License
- CC BY-NC 4.0
Frequently Asked Questions
How much VRAM does SOLAR 10.7B Instruct v1.0 need?
SOLAR 10.7B Instruct v1.0 requires 7.1 GB of VRAM at Q4_K_M, or 22.2 GB at FP16.
VRAM = Weights + KV Cache + Overhead
Weights = 10.7B × 4.8 bits ÷ 8 = 6.4 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 ≈ 1.1 GB (at full 4K context)
VRAM usage by quantization
What's the best quantization for SOLAR 10.7B Instruct v1.0?
For SOLAR 10.7B Instruct v1.0, Q4_K_M (7.1 GB) offers the best balance of quality and VRAM usage. Q5_K_M (8.3 GB) provides better quality if you have the VRAM. The smallest option is Q2_K at 5.3 GB.
VRAM requirement by quantization
★ Recommended — best balance of quality and VRAM usage.
Can I run SOLAR 10.7B Instruct v1.0 on a Mac?
Yes — Mac Mini M2 Pro (16 GB) and 133 other Macs can run SOLAR 10.7B Instruct v1.0. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 7.1 GB of usable unified memory (RAM minus macOS overhead).
Can I run SOLAR 10.7B Instruct v1.0 locally?
Yes — SOLAR 10.7B Instruct v1.0 can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is SOLAR 10.7B Instruct v1.0?
At Q4_K_M, SOLAR 10.7B Instruct v1.0 can reach ~672 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~92 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 ÷ 7.14 × 0.65 = ~728 tok/s
Estimated speed at Q4_K_M (7.1 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of SOLAR 10.7B Instruct v1.0?
At Q4_K_M, the download is about 6.44 GB. The full-precision FP16 version is 21.46 GB. The smallest option (Q2_K) is 4.56 GB.
Which GPUs can run SOLAR 10.7B Instruct v1.0?
52 consumer GPUs can run SOLAR 10.7B Instruct v1.0 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. 38 GPUs have plenty of headroom for comfortable inference.
Which devices can run SOLAR 10.7B Instruct v1.0?
151 devices with unified memory can run SOLAR 10.7B Instruct v1.0 at Q4_K_M (7.1 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.