Quick start

ollama run granite3-dense:latest

Runnable variants

Memory is a rough file-size estimate; context length, runtime and GPU offload change it.

12 of 33 matching variants shown.33 of 33 variants shown.

Tag	Type	Inputs	Digest	Download	Parameters	Quantization	Context	Est. memory	Run
2b	Typelocal	Inputstext	Digest5c2e6f3112f4	Download1.6 GB	Parameters2b	QuantizationUnknown	Context4K	Est. memory≈ 3 GB	`ollama run granite3-dense:2b`
2b-instruct-fp16	Typelocal	Inputstext	Digest24e5c0a35f78	Download5.3 GB	Parameters2b	QuantizationFP16	Context4K	Est. memory≈ 7 GB	`ollama run granite3-dense:2b-instruct-fp16`
2b-instruct-q2_K	Typelocal	Inputstext	Digest2261d92d6175	Download1.0 GB	Parameters2b	QuantizationQ2_K	Context4K	Est. memory≈ 2.5 GB	`ollama run granite3-dense:2b-instruct-q2_K`
2b-instruct-q3_K_L	Typelocal	Inputstext	Digest201ccf1522e8	Download1.4 GB	Parameters2b	QuantizationQ3_K_L	Context4K	Est. memory≈ 3 GB	`ollama run granite3-dense:2b-instruct-q3_K_L`
2b-instruct-q3_K_M	Typelocal	Inputstext	Digestc08ae454856c	Download1.3 GB	Parameters2b	QuantizationQ3_K_M	Context4K	Est. memory≈ 2.5 GB	`ollama run granite3-dense:2b-instruct-q3_K_M`
2b-instruct-q3_K_S	Typelocal	Inputstext	Digest7c6af70b9f11	Download1.2 GB	Parameters2b	QuantizationQ3_K_S	Context4K	Est. memory≈ 2.5 GB	`ollama run granite3-dense:2b-instruct-q3_K_S`
2b-instruct-q4_0	Typelocal	Inputstext	Digestdd4a07b9e1b7	Download1.5 GB	Parameters2b	QuantizationQ4_0	Context4K	Est. memory≈ 3 GB	`ollama run granite3-dense:2b-instruct-q4_0`
2b-instruct-q4_1	Typelocal	Inputstext	Digeste12ad0fad2ad	Download1.7 GB	Parameters2b	QuantizationQ4_1	Context4K	Est. memory≈ 3 GB	`ollama run granite3-dense:2b-instruct-q4_1`
2b-instruct-q4_K_M	Typelocal	Inputstext	Digest5c2e6f3112f4	Download1.6 GB	Parameters2b	QuantizationQ4_K_M	Context4K	Est. memory≈ 3 GB	`ollama run granite3-dense:2b-instruct-q4_K_M`
2b-instruct-q4_K_S	Typelocal	Inputstext	Digest0f87f54eb387	Download1.5 GB	Parameters2b	QuantizationQ4_K_S	Context4K	Est. memory≈ 3 GB	`ollama run granite3-dense:2b-instruct-q4_K_S`
2b-instruct-q5_0	Typelocal	Inputstext	Digestdb3fbcdec89b	Download1.8 GB	Parameters2b	QuantizationQ5_0	Context4K	Est. memory≈ 3.5 GB	`ollama run granite3-dense:2b-instruct-q5_0`
2b-instruct-q5_1	Typelocal	Inputstext	Digestbc5ed577d64f	Download2.0 GB	Parameters2b	QuantizationQ5_1	Context4K	Est. memory≈ 3.5 GB	`ollama run granite3-dense:2b-instruct-q5_1`
2b-instruct-q5_K_M	Typelocal	Inputstext	Digestdcace13d6d34	Download1.9 GB	Parameters2b	QuantizationQ5_K_M	Context4K	Est. memory≈ 3.5 GB	`ollama run granite3-dense:2b-instruct-q5_K_M`
2b-instruct-q5_K_S	Typelocal	Inputstext	Digestbe9d8efefd30	Download1.8 GB	Parameters2b	QuantizationQ5_K_S	Context4K	Est. memory≈ 3.5 GB	`ollama run granite3-dense:2b-instruct-q5_K_S`
2b-instruct-q6_K	Typelocal	Inputstext	Digest7f374efdad9b	Download2.2 GB	Parameters2b	QuantizationQ6_K	Context4K	Est. memory≈ 3.5 GB	`ollama run granite3-dense:2b-instruct-q6_K`
2b-instruct-q8_0	Typelocal	Inputstext	Digestf35cd2e1e433	Download2.8 GB	Parameters2b	QuantizationQ8_0	Context4K	Est. memory≈ 4.5 GB	`ollama run granite3-dense:2b-instruct-q8_0`
8b	Typelocal	Inputstext	Digest199456d876ee	Download4.9 GB	Parameters8b	QuantizationUnknown	Context4K	Est. memory≈ 6.5 GB	`ollama run granite3-dense:8b`
8b-instruct-fp16	Typelocal	Inputstext	Digestd3a4062e339e	Download16 GB	Parameters8b	QuantizationFP16	Context4K	Est. memory≈ 19 GB	`ollama run granite3-dense:8b-instruct-fp16`
8b-instruct-q2_K	Typelocal	Inputstext	Digest2ea57f8bc8fd	Download3.1 GB	Parameters8b	QuantizationQ2_K	Context4K	Est. memory≈ 4.5 GB	`ollama run granite3-dense:8b-instruct-q2_K`
8b-instruct-q3_K_L	Typelocal	Inputstext	Digestc74b04a2513f	Download4.3 GB	Parameters8b	QuantizationQ3_K_L	Context4K	Est. memory≈ 6 GB	`ollama run granite3-dense:8b-instruct-q3_K_L`
8b-instruct-q3_K_M	Typelocal	Inputstext	Digestc1a94d47b25c	Download4.0 GB	Parameters8b	QuantizationQ3_K_M	Context4K	Est. memory≈ 5.5 GB	`ollama run granite3-dense:8b-instruct-q3_K_M`
8b-instruct-q3_K_S	Typelocal	Inputstext	Digest4c5f0dba11a8	Download3.6 GB	Parameters8b	QuantizationQ3_K_S	Context4K	Est. memory≈ 5.5 GB	`ollama run granite3-dense:8b-instruct-q3_K_S`
8b-instruct-q4_0	Typelocal	Inputstext	Digest16473222519f	Download4.7 GB	Parameters8b	QuantizationQ4_0	Context4K	Est. memory≈ 6.5 GB	`ollama run granite3-dense:8b-instruct-q4_0`
8b-instruct-q4_1	Typelocal	Inputstext	Digest8a47472a11f2	Download5.1 GB	Parameters8b	QuantizationQ4_1	Context4K	Est. memory≈ 7 GB	`ollama run granite3-dense:8b-instruct-q4_1`
8b-instruct-q4_K_M	Typelocal	Inputstext	Digest199456d876ee	Download4.9 GB	Parameters8b	QuantizationQ4_K_M	Context4K	Est. memory≈ 6.5 GB	`ollama run granite3-dense:8b-instruct-q4_K_M`
8b-instruct-q4_K_S	Typelocal	Inputstext	Digest8fc1ce76b8f2	Download4.7 GB	Parameters8b	QuantizationQ4_K_S	Context4K	Est. memory≈ 6.5 GB	`ollama run granite3-dense:8b-instruct-q4_K_S`
8b-instruct-q5_0	Typelocal	Inputstext	Digest449cddb1864d	Download5.6 GB	Parameters8b	QuantizationQ5_0	Context4K	Est. memory≈ 7.5 GB	`ollama run granite3-dense:8b-instruct-q5_0`
8b-instruct-q5_1	Typelocal	Inputstext	Digestae4ab94be005	Download6.1 GB	Parameters8b	QuantizationQ5_1	Context4K	Est. memory≈ 8 GB	`ollama run granite3-dense:8b-instruct-q5_1`
8b-instruct-q5_K_M	Typelocal	Inputstext	Digeste85361d1973c	Download5.8 GB	Parameters8b	QuantizationQ5_K_M	Context4K	Est. memory≈ 7.5 GB	`ollama run granite3-dense:8b-instruct-q5_K_M`
8b-instruct-q5_K_S	Typelocal	Inputstext	Digest49055e666925	Download5.6 GB	Parameters8b	QuantizationQ5_K_S	Context4K	Est. memory≈ 7.5 GB	`ollama run granite3-dense:8b-instruct-q5_K_S`
8b-instruct-q6_K	Typelocal	Inputstext	Digestec258d53940b	Download6.7 GB	Parameters8b	QuantizationQ6_K	Context4K	Est. memory≈ 8.5 GB	`ollama run granite3-dense:8b-instruct-q6_K`
8b-instruct-q8_0	Typelocal	Inputstext	Digestf1466e70a7be	Download8.7 GB	Parameters8b	QuantizationQ8_0	Context4K	Est. memory≈ 11 GB	`ollama run granite3-dense:8b-instruct-q8_0`
latestdefault	Typelocal	Inputstext	Digest5c2e6f3112f4	Download1.6 GB	ParametersUnknown	QuantizationUnknown	Context4K	Est. memory≈ 3 GB	`ollama run granite3-dense:latest`

Source-listed applications

Claude CodeHermes AgentOpenClawOpenCode

Related models

qwen3.5local

Qwen 3.5 is a family of open-source multimodal models that delivers exceptional utility and performance.

granite3.1-denselocal

The IBM Granite 2B and 8B models are text-only dense LLMs trained on over 12 trillion tokens of data, demonstrated significant improvements over their predecessors in performance and speed in IBM’s initial testing.

bonsai-2local

Bonsai 2 from Prism ML is a family of vision models built for ternary weights – smaller models with close to full-precision quality.

deepseek-r1local

DeepSeek-R1 is a family of open reasoning models with performance approaching that of leading models, such as O3 and Gemini 2.5 Pro.