Can I run Olmo 3 1025 7B locally?
Olmo 3 1025 7B by allenai has 7.3B parameters. At 4-bit with an 8K context it needs about 7.3 GB of GPU memory. It fits on 65 of 65 GPUs and Macs we track. See all models that fit in 8 GB of VRAM.
Memory needed by quant and context length
GB of GPU memory = weights + KV cache (FP16) + runtime overhead. Weights use real file sizes from lmstudio-community/Olmo-3-1025-7B-GGUF where available (marked •), otherwise parameter count × bits per weight. How we calculate.
| Quant | Quality | 4K ctx | 8K ctx | 32K ctx |
|---|---|---|---|---|
| 2-bit (Q2_K) | rough | 5.1 GB | 5.6 GB | 8.6 GB |
| 3-bit (Q3_K_M) | usable | 5.9 GB | 6.4 GB | 9.4 GB |
| 4-bit (Q4_K_M) • | good, most popular | 6.8 GB | 7.3 GB | 10.3 GB |
| 5-bit (Q5_K_M) | very good | 7.4 GB | 7.9 GB | 10.9 GB |
| 6-bit (Q6_K) • | near lossless | 8.2 GB | 8.7 GB | 11.7 GB |
| 8-bit (Q8_0) • | effectively lossless | 9.8 GB | 10.3 GB | 13.3 GB |
| 16-bit (F16/BF16) | original weights | 16.2 GB | 16.7 GB | 19.7 GB |
Which GPUs and Macs run Olmo 3 1025 7B?
Best quality quant that fits at 8K context. "Tight" = within 10% of usable memory. Speed is a ceiling from memory bandwidth (real speed is typically 50–80% of it).
| Hardware | Memory | Fits? | Best quant | Needs | Max tok/s (est.) | |
|---|---|---|---|---|---|---|
| RTX 3060 12GB | 12 GB | Fits | Q6_K | 8.7 GB | 60 | details |
| RTX 4060 Ti 16GB | 16 GB | Fits | Q8_0 | 10.3 GB | 37 | details |
| RTX 3090 | 24 GB | Fits | F16 | 16.7 GB | 64 | details |
| RTX 4090 | 24 GB | Fits | F16 | 16.7 GB | 69 | details |
| RTX 5090 | 32 GB | Fits | F16 | 16.7 GB | 123 | details |
| RTX 5070 Ti | 16 GB | Fits | Q8_0 | 10.3 GB | 115 | details |
| RX 7900 XTX | 24 GB | Fits | F16 | 16.7 GB | 66 | details |
| Mac M4 Pro 48GB | 48 GB | Fits | F16 | 16.7 GB | 19 | details |
| Mac M4 Max 128GB | 128 GB | Fits | F16 | 16.7 GB | 37 | details |
| Mac M3 Ultra 512GB | 512 GB | Fits | F16 | 16.7 GB | 56 | details |
| DGX Spark | 128 GB | Fits | F16 | 16.7 GB | 19 | details |
| RTX 4070 | 12 GB | Fits | Q6_K | 8.7 GB | 84 | details |
| RTX 5060 Ti 16GB | 16 GB | Fits | Q8_0 | 10.3 GB | 58 | details |
| RTX 4060 | 8 GB | Fits | Q3_K_M | 6.4 GB | 76 | details |
| Arc B580 | 12 GB | Fits | Q6_K | 8.7 GB | 76 | details |
| RTX 3060 Ti | 8 GB | Fits | Q3_K_M | 6.4 GB | 126 | details |
| RTX 3070 | 8 GB | Fits | Q3_K_M | 6.4 GB | 126 | details |
| RTX 3080 10GB | 10 GB | Fits | Q5_K_M | 7.9 GB | 146 | details |
| RTX 3080 12GB | 12 GB | Fits | Q6_K | 8.7 GB | 152 | details |
| RTX 3090 Ti | 24 GB | Fits | F16 | 16.7 GB | 69 | details |
| RTX 4060 Ti 8GB | 8 GB | Fits | Q3_K_M | 6.4 GB | 81 | details |
| RTX 4070 Super | 12 GB | Fits | Q6_K | 8.7 GB | 84 | details |
| RTX 4070 Ti | 12 GB | Fits | Q6_K | 8.7 GB | 84 | details |
| RTX 4070 Ti Super | 16 GB | Fits | Q8_0 | 10.3 GB | 87 | details |
| RTX 4080 | 16 GB | Fits | Q8_0 | 10.3 GB | 92 | details |
| RTX 4080 Super | 16 GB | Fits | Q8_0 | 10.3 GB | 95 | details |
| RTX 5060 | 8 GB | Fits | Q3_K_M | 6.4 GB | 126 | details |
| RTX 5070 | 12 GB | Fits | Q6_K | 8.7 GB | 112 | details |
| RTX 5080 | 16 GB | Fits | Q8_0 | 10.3 GB | 124 | details |
| 2x RTX 3090 | 2×24 GB | Fits | F16 | 16.7 GB | 64 | details |
| 2x RTX 4090 | 2×24 GB | Fits | F16 | 16.7 GB | 69 | details |
| 2x RTX 5090 | 2×32 GB | Fits | F16 | 16.7 GB | 123 | details |
| RTX A6000 | 48 GB | Fits | F16 | 16.7 GB | 53 | details |
| RTX 6000 Ada | 48 GB | Fits | F16 | 16.7 GB | 66 | details |
| RTX PRO 6000 Blackwell | 96 GB | Fits | F16 | 16.7 GB | 123 | details |
| L40S | 48 GB | Fits | F16 | 16.7 GB | 59 | details |
| A100 40GB | 40 GB | Fits | F16 | 16.7 GB | 107 | details |
| A100 80GB | 80 GB | Fits | F16 | 16.7 GB | 140 | details |
| H100 PCIe | 80 GB | Fits | F16 | 16.7 GB | 137 | details |
| H100 SXM | 80 GB | Fits | F16 | 16.7 GB | 230 | details |
| H200 | 141 GB | Fits | F16 | 16.7 GB | 329 | details |
| RX 7600 | 8 GB | Fits | Q3_K_M | 6.4 GB | 81 | details |
| RX 7700 XT | 12 GB | Fits | Q6_K | 8.7 GB | 72 | details |
| RX 7800 XT | 16 GB | Fits | Q8_0 | 10.3 GB | 80 | details |
| RX 7900 GRE | 16 GB | Fits | Q8_0 | 10.3 GB | 74 | details |
| RX 7900 XT | 20 GB | Fits | F16 | 16.7 GB | 55 | details |
| RX 9070 | 16 GB | Fits | Q8_0 | 10.3 GB | 82 | details |
| RX 9070 XT | 16 GB | Fits | Q8_0 | 10.3 GB | 82 | details |
| Ryzen AI Max+ 395 128GB | 128 GB | Fits | F16 | 16.7 GB | 18 | details |
| Arc A770 16GB | 16 GB | Fits | Q8_0 | 10.3 GB | 72 | details |
| Mac M1 16GB | 16 GB | Fits | Q6_K | 8.7 GB | 11 | details |
| Mac M2 24GB | 24 GB | Fits | Q8_0 | 10.3 GB | 13 | details |
| Mac M3 24GB | 24 GB | Fits | Q8_0 | 10.3 GB | 13 | details |
| Mac M4 16GB | 16 GB | Fits | Q6_K | 8.7 GB | 20 | details |
| Mac M4 32GB | 32 GB | Fits | F16 | 16.7 GB | 8.2 | details |
| Mac M1 Pro 32GB | 32 GB | Fits | F16 | 16.7 GB | 14 | details |
| Mac M3 Pro 36GB | 36 GB | Fits | F16 | 16.7 GB | 10 | details |
| Mac M4 Pro 24GB | 24 GB | Fits | Q8_0 | 10.3 GB | 35 | details |
| Mac M4 Pro 64GB | 64 GB | Fits | F16 | 16.7 GB | 19 | details |
| Mac M1 Max 64GB | 64 GB | Fits | F16 | 16.7 GB | 27 | details |
| Mac M2 Max 96GB | 96 GB | Fits | F16 | 16.7 GB | 27 | details |
| Mac M3 Max 128GB | 128 GB | Fits | F16 | 16.7 GB | 27 | details |
| Mac M4 Max 64GB | 64 GB | Fits | F16 | 16.7 GB | 37 | details |
| Mac M2 Ultra 192GB | 192 GB | Fits | F16 | 16.7 GB | 55 | details |
| Mac M3 Ultra 256GB | 256 GB | Fits | F16 | 16.7 GB | 56 | details |
Questions
How much VRAM does Olmo 3 1025 7B need?
About 7.3 GB at 4-bit (Q4_K_M) with an 8K context: 4.2 GB of weights (real GGUF file size), 2.5 GB of KV cache and 0.6 GB of runtime overhead. At 8-bit it needs about 10.3 GB.
What is the smallest GPU that runs Olmo 3 1025 7B?
At Q4_K_M and 8K context, the smallest single consumer GPU in our table that fits it is the RTX 3080 10GB (10 GB).
Can I run Olmo 3 1025 7B on a Mac?
Yes. At Q4_K_M with 8K context it fits on a Mac M1 16GB or larger, using the default macOS GPU memory limit.
Similar models
Same family or a similar parameter count, with the VRAM each needs at Q4_K_M and 8K context.
Olmo 3 7B Instruct
7.3B · 7.3 GB at 4-bitOlmo 3 7B Think
7.3B · 7.3 GB at 4-bitOLMoE 1B 7B 0924
6.9B · 5.4 GB at 4-bitMistral 7B v0.1
7.2B · 5.5 GB at 4-bitMistral 7B Instruct v0.2
7.2B · 47.7 GB at 4-bitMistral 7B Instruct v0.1
7.2B · 5.5 GB at 4-bitModel facts
| Hugging Face repo | allenai/Olmo-3-1025-7B |
| Parameters | 7,298,011,136 (7.3B) |
| Layers / hidden size | 32 / 4096 |
| Attention | MHA + sliding window; 32 heads, 32 KV heads, head dim 128; sliding window 4096 |
| KV cache per token (FP16) | 512.0 KiB (sliding layers capped at the window) |
| Max context | 65,536 tokens |
| License | apache-2.0 |
| Released on HF | Sep 12, 2025 |
| Downloads (30d) / likes | 103,989 / 93 |
Sources: Hugging Face model API · config.json (allenai/Olmo-3-1025-7B) · GGUF files · GPU specs from vendor pages (list). Data fetched Nov 17, 2026. Spot an error? Report it.