Models / nemotron_h

NVIDIA-Nemotron-3-Nano-30B-A3B-NVFP4

Reconstructed from its own config.json with no weights read. 497K downloads on Hugging Face.

Our count against the checkpoint

The left number comes from the graph. The right one is the number of scalars in the published weight files. Nothing on this page was tuned to make them agree.

Derived from structure

102B

102,337,027,328 parameters

In the published checkpoint

18.24B

18,237,772,608 scalars · safetensors.total, read 2026-08-24

Delta

+461%

custom-code This repository ships its own modeling code (`auto_map`, e.g. `configuration_nemotron_h.py`), so `config.json` names a class in the repo rather than an architecture `transformers` defines. The graph below is what those config keys mean under `transformers` semantics, which is not necessarily what the repo's own file builds. A gap here is a statement about what we read, not about the checkpoint.

What it costs to run

Cost is a roofline estimate on the priced GPU for 10 epochs at batch 32 over 50,000 samples (assumed; no dataset attached). GPU fit is fp32 weights plus gradients plus two Adam moments (16 bytes per parameter) with 1.3x headroom; activations are not included and grow with batch size.

Layers

316

Will it forward-pass

Yes

Priced on

A10G (24GB)

Est. one run

$616838.54

Card	Memory
T4 (16GB)	weights + activations	does not fit
A100 (40GB)	weights + activations	does not fit
H100 (80GB)	weights + activations	does not fit

Structure

318 nodes. Output shapes are propagated from the input shape, batch dimension excluded.

	Layer	Type	Output shape
1	Input	Input	1 × 262144
2	Embedding	Embedding	1 × 262144 × 2688
3	RoPE	RoPE	1 × 262144 × 2688
4	LayerNorm_1_1	LayerNorm	1 × 262144 × 2688
5	Attention_1	Grouped Query Attn	1 × 262144 × 2688
6	Add_1_attn	Add	1 × 262144 × 2688
7	LayerNorm_1_2	LayerNorm	1 × 262144 × 2688
8	MoE_1	Shared-Expert MoE	1 × 262144 × 2688
9	Add_1_ffn	Add	1 × 262144 × 2688
10	LayerNorm_2_1	LayerNorm	1 × 262144 × 2688
11	Attention_2	Grouped Query Attn	1 × 262144 × 2688
12	Add_2_attn	Add	1 × 262144 × 2688
13	LayerNorm_2_2	LayerNorm	1 × 262144 × 2688
14	MoE_2	Shared-Expert MoE	1 × 262144 × 2688
15	Add_2_ffn	Add	1 × 262144 × 2688
16	LayerNorm_3_1	LayerNorm	1 × 262144 × 2688
17	Attention_3	Grouped Query Attn	1 × 262144 × 2688
18	Add_3_attn	Add	1 × 262144 × 2688
19	LayerNorm_3_2	LayerNorm	1 × 262144 × 2688
20	MoE_3	Shared-Expert MoE	1 × 262144 × 2688
21	Add_3_ffn	Add	1 × 262144 × 2688
22	LayerNorm_4_1	LayerNorm	1 × 262144 × 2688
23	Attention_4	Grouped Query Attn	1 × 262144 × 2688
24	Add_4_attn	Add	1 × 262144 × 2688
25	LayerNorm_4_2	LayerNorm	1 × 262144 × 2688
26	MoE_4	Shared-Expert MoE	1 × 262144 × 2688
27	Add_4_ffn	Add	1 × 262144 × 2688
28	LayerNorm_5_1	LayerNorm	1 × 262144 × 2688
29	Attention_5	Grouped Query Attn	1 × 262144 × 2688
30	Add_5_attn	Add	1 × 262144 × 2688
31	LayerNorm_5_2	LayerNorm	1 × 262144 × 2688
32	MoE_5	Shared-Expert MoE	1 × 262144 × 2688
33	Add_5_ffn	Add	1 × 262144 × 2688
34	LayerNorm_6_1	LayerNorm	1 × 262144 × 2688
35	Attention_6	Grouped Query Attn	1 × 262144 × 2688
36	Add_6_attn	Add	1 × 262144 × 2688
37	LayerNorm_6_2	LayerNorm	1 × 262144 × 2688
38	MoE_6	Shared-Expert MoE	1 × 262144 × 2688
39	Add_6_ffn	Add	1 × 262144 × 2688
40	LayerNorm_7_1	LayerNorm	1 × 262144 × 2688
41	Attention_7	Grouped Query Attn	1 × 262144 × 2688
42	Add_7_attn	Add	1 × 262144 × 2688
43	LayerNorm_7_2	LayerNorm	1 × 262144 × 2688
44	MoE_7	Shared-Expert MoE	1 × 262144 × 2688
45	Add_7_ffn	Add	1 × 262144 × 2688
46	LayerNorm_8_1	LayerNorm	1 × 262144 × 2688
47	Attention_8	Grouped Query Attn	1 × 262144 × 2688
48	Add_8_attn	Add	1 × 262144 × 2688
49	LayerNorm_8_2	LayerNorm	1 × 262144 × 2688
50	MoE_8	Shared-Expert MoE	1 × 262144 × 2688
51	Add_8_ffn	Add	1 × 262144 × 2688
52	LayerNorm_9_1	LayerNorm	1 × 262144 × 2688
53	Attention_9	Grouped Query Attn	1 × 262144 × 2688
54	Add_9_attn	Add	1 × 262144 × 2688
55	LayerNorm_9_2	LayerNorm	1 × 262144 × 2688
56	MoE_9	Shared-Expert MoE	1 × 262144 × 2688
57	Add_9_ffn	Add	1 × 262144 × 2688
58	LayerNorm_10_1	LayerNorm	1 × 262144 × 2688
59	Attention_10	Grouped Query Attn	1 × 262144 × 2688
60	Add_10_attn	Add	1 × 262144 × 2688
61	LayerNorm_10_2	LayerNorm	1 × 262144 × 2688
62	MoE_10	Shared-Expert MoE	1 × 262144 × 2688
63	Add_10_ffn	Add	1 × 262144 × 2688
64	LayerNorm_11_1	LayerNorm	1 × 262144 × 2688
65	Attention_11	Grouped Query Attn	1 × 262144 × 2688
66	Add_11_attn	Add	1 × 262144 × 2688
67	LayerNorm_11_2	LayerNorm	1 × 262144 × 2688
68	MoE_11	Shared-Expert MoE	1 × 262144 × 2688
69	Add_11_ffn	Add	1 × 262144 × 2688
70	LayerNorm_12_1	LayerNorm	1 × 262144 × 2688
71	Attention_12	Grouped Query Attn	1 × 262144 × 2688
72	Add_12_attn	Add	1 × 262144 × 2688
73	LayerNorm_12_2	LayerNorm	1 × 262144 × 2688
74	MoE_12	Shared-Expert MoE	1 × 262144 × 2688
75	Add_12_ffn	Add	1 × 262144 × 2688
76	LayerNorm_13_1	LayerNorm	1 × 262144 × 2688
77	Attention_13	Grouped Query Attn	1 × 262144 × 2688
78	Add_13_attn	Add	1 × 262144 × 2688
79	LayerNorm_13_2	LayerNorm	1 × 262144 × 2688
80	MoE_13	Shared-Expert MoE	1 × 262144 × 2688
81	Add_13_ffn	Add	1 × 262144 × 2688
82	LayerNorm_14_1	LayerNorm	1 × 262144 × 2688
83	Attention_14	Grouped Query Attn	1 × 262144 × 2688
84	Add_14_attn	Add	1 × 262144 × 2688
85	LayerNorm_14_2	LayerNorm	1 × 262144 × 2688
86	MoE_14	Shared-Expert MoE	1 × 262144 × 2688
87	Add_14_ffn	Add	1 × 262144 × 2688
88	LayerNorm_15_1	LayerNorm	1 × 262144 × 2688
89	Attention_15	Grouped Query Attn	1 × 262144 × 2688
90	Add_15_attn	Add	1 × 262144 × 2688
91	LayerNorm_15_2	LayerNorm	1 × 262144 × 2688
92	MoE_15	Shared-Expert MoE	1 × 262144 × 2688
93	Add_15_ffn	Add	1 × 262144 × 2688
94	LayerNorm_16_1	LayerNorm	1 × 262144 × 2688
95	Attention_16	Grouped Query Attn	1 × 262144 × 2688
96	Add_16_attn	Add	1 × 262144 × 2688
97	LayerNorm_16_2	LayerNorm	1 × 262144 × 2688
98	MoE_16	Shared-Expert MoE	1 × 262144 × 2688
99	Add_16_ffn	Add	1 × 262144 × 2688
100	LayerNorm_17_1	LayerNorm	1 × 262144 × 2688
101	Attention_17	Grouped Query Attn	1 × 262144 × 2688
102	Add_17_attn	Add	1 × 262144 × 2688
103	LayerNorm_17_2	LayerNorm	1 × 262144 × 2688
104	MoE_17	Shared-Expert MoE	1 × 262144 × 2688
105	Add_17_ffn	Add	1 × 262144 × 2688
106	LayerNorm_18_1	LayerNorm	1 × 262144 × 2688
107	Attention_18	Grouped Query Attn	1 × 262144 × 2688
108	Add_18_attn	Add	1 × 262144 × 2688
109	LayerNorm_18_2	LayerNorm	1 × 262144 × 2688
110	MoE_18	Shared-Expert MoE	1 × 262144 × 2688
111	Add_18_ffn	Add	1 × 262144 × 2688
112	LayerNorm_19_1	LayerNorm	1 × 262144 × 2688
113	Attention_19	Grouped Query Attn	1 × 262144 × 2688
114	Add_19_attn	Add	1 × 262144 × 2688
115	LayerNorm_19_2	LayerNorm	1 × 262144 × 2688
116	MoE_19	Shared-Expert MoE	1 × 262144 × 2688
117	Add_19_ffn	Add	1 × 262144 × 2688
118	LayerNorm_20_1	LayerNorm	1 × 262144 × 2688
119	Attention_20	Grouped Query Attn	1 × 262144 × 2688
120	Add_20_attn	Add	1 × 262144 × 2688
121	LayerNorm_20_2	LayerNorm	1 × 262144 × 2688
122	MoE_20	Shared-Expert MoE	1 × 262144 × 2688
123	Add_20_ffn	Add	1 × 262144 × 2688
124	LayerNorm_21_1	LayerNorm	1 × 262144 × 2688
125	Attention_21	Grouped Query Attn	1 × 262144 × 2688
126	Add_21_attn	Add	1 × 262144 × 2688
127	LayerNorm_21_2	LayerNorm	1 × 262144 × 2688
128	MoE_21	Shared-Expert MoE	1 × 262144 × 2688
129	Add_21_ffn	Add	1 × 262144 × 2688
130	LayerNorm_22_1	LayerNorm	1 × 262144 × 2688
131	Attention_22	Grouped Query Attn	1 × 262144 × 2688
132	Add_22_attn	Add	1 × 262144 × 2688
133	LayerNorm_22_2	LayerNorm	1 × 262144 × 2688
134	MoE_22	Shared-Expert MoE	1 × 262144 × 2688
135	Add_22_ffn	Add	1 × 262144 × 2688
136	LayerNorm_23_1	LayerNorm	1 × 262144 × 2688
137	Attention_23	Grouped Query Attn	1 × 262144 × 2688
138	Add_23_attn	Add	1 × 262144 × 2688
139	LayerNorm_23_2	LayerNorm	1 × 262144 × 2688
140	MoE_23	Shared-Expert MoE	1 × 262144 × 2688
141	Add_23_ffn	Add	1 × 262144 × 2688
142	LayerNorm_24_1	LayerNorm	1 × 262144 × 2688
143	Attention_24	Grouped Query Attn	1 × 262144 × 2688
144	Add_24_attn	Add	1 × 262144 × 2688
145	LayerNorm_24_2	LayerNorm	1 × 262144 × 2688
146	MoE_24	Shared-Expert MoE	1 × 262144 × 2688
147	Add_24_ffn	Add	1 × 262144 × 2688
148	LayerNorm_25_1	LayerNorm	1 × 262144 × 2688
149	Attention_25	Grouped Query Attn	1 × 262144 × 2688
150	Add_25_attn	Add	1 × 262144 × 2688
151	LayerNorm_25_2	LayerNorm	1 × 262144 × 2688
152	MoE_25	Shared-Expert MoE	1 × 262144 × 2688
153	Add_25_ffn	Add	1 × 262144 × 2688
154	LayerNorm_26_1	LayerNorm	1 × 262144 × 2688
155	Attention_26	Grouped Query Attn	1 × 262144 × 2688
156	Add_26_attn	Add	1 × 262144 × 2688
157	LayerNorm_26_2	LayerNorm	1 × 262144 × 2688
158	MoE_26	Shared-Expert MoE	1 × 262144 × 2688
159	Add_26_ffn	Add	1 × 262144 × 2688
160	LayerNorm_27_1	LayerNorm	1 × 262144 × 2688
161	Attention_27	Grouped Query Attn	1 × 262144 × 2688
162	Add_27_attn	Add	1 × 262144 × 2688
163	LayerNorm_27_2	LayerNorm	1 × 262144 × 2688
164	MoE_27	Shared-Expert MoE	1 × 262144 × 2688
165	Add_27_ffn	Add	1 × 262144 × 2688
166	LayerNorm_28_1	LayerNorm	1 × 262144 × 2688
167	Attention_28	Grouped Query Attn	1 × 262144 × 2688
168	Add_28_attn	Add	1 × 262144 × 2688
169	LayerNorm_28_2	LayerNorm	1 × 262144 × 2688
170	MoE_28	Shared-Expert MoE	1 × 262144 × 2688
171	Add_28_ffn	Add	1 × 262144 × 2688
172	LayerNorm_29_1	LayerNorm	1 × 262144 × 2688
173	Attention_29	Grouped Query Attn	1 × 262144 × 2688
174	Add_29_attn	Add	1 × 262144 × 2688
175	LayerNorm_29_2	LayerNorm	1 × 262144 × 2688
176	MoE_29	Shared-Expert MoE	1 × 262144 × 2688
177	Add_29_ffn	Add	1 × 262144 × 2688
178	LayerNorm_30_1	LayerNorm	1 × 262144 × 2688
179	Attention_30	Grouped Query Attn	1 × 262144 × 2688
180	Add_30_attn	Add	1 × 262144 × 2688
181	LayerNorm_30_2	LayerNorm	1 × 262144 × 2688
182	MoE_30	Shared-Expert MoE	1 × 262144 × 2688
183	Add_30_ffn	Add	1 × 262144 × 2688
184	LayerNorm_31_1	LayerNorm	1 × 262144 × 2688
185	Attention_31	Grouped Query Attn	1 × 262144 × 2688
186	Add_31_attn	Add	1 × 262144 × 2688
187	LayerNorm_31_2	LayerNorm	1 × 262144 × 2688
188	MoE_31	Shared-Expert MoE	1 × 262144 × 2688
189	Add_31_ffn	Add	1 × 262144 × 2688
190	LayerNorm_32_1	LayerNorm	1 × 262144 × 2688
191	Attention_32	Grouped Query Attn	1 × 262144 × 2688
192	Add_32_attn	Add	1 × 262144 × 2688
193	LayerNorm_32_2	LayerNorm	1 × 262144 × 2688
194	MoE_32	Shared-Expert MoE	1 × 262144 × 2688
195	Add_32_ffn	Add	1 × 262144 × 2688
196	LayerNorm_33_1	LayerNorm	1 × 262144 × 2688
197	Attention_33	Grouped Query Attn	1 × 262144 × 2688
198	Add_33_attn	Add	1 × 262144 × 2688
199	LayerNorm_33_2	LayerNorm	1 × 262144 × 2688
200	MoE_33	Shared-Expert MoE	1 × 262144 × 2688
201	Add_33_ffn	Add	1 × 262144 × 2688
202	LayerNorm_34_1	LayerNorm	1 × 262144 × 2688
203	Attention_34	Grouped Query Attn	1 × 262144 × 2688
204	Add_34_attn	Add	1 × 262144 × 2688
205	LayerNorm_34_2	LayerNorm	1 × 262144 × 2688
206	MoE_34	Shared-Expert MoE	1 × 262144 × 2688
207	Add_34_ffn	Add	1 × 262144 × 2688
208	LayerNorm_35_1	LayerNorm	1 × 262144 × 2688
209	Attention_35	Grouped Query Attn	1 × 262144 × 2688
210	Add_35_attn	Add	1 × 262144 × 2688
211	LayerNorm_35_2	LayerNorm	1 × 262144 × 2688
212	MoE_35	Shared-Expert MoE	1 × 262144 × 2688
213	Add_35_ffn	Add	1 × 262144 × 2688
214	LayerNorm_36_1	LayerNorm	1 × 262144 × 2688
215	Attention_36	Grouped Query Attn	1 × 262144 × 2688
216	Add_36_attn	Add	1 × 262144 × 2688
217	LayerNorm_36_2	LayerNorm	1 × 262144 × 2688
218	MoE_36	Shared-Expert MoE	1 × 262144 × 2688
219	Add_36_ffn	Add	1 × 262144 × 2688
220	LayerNorm_37_1	LayerNorm	1 × 262144 × 2688
221	Attention_37	Grouped Query Attn	1 × 262144 × 2688
222	Add_37_attn	Add	1 × 262144 × 2688
223	LayerNorm_37_2	LayerNorm	1 × 262144 × 2688
224	MoE_37	Shared-Expert MoE	1 × 262144 × 2688
225	Add_37_ffn	Add	1 × 262144 × 2688
226	LayerNorm_38_1	LayerNorm	1 × 262144 × 2688
227	Attention_38	Grouped Query Attn	1 × 262144 × 2688
228	Add_38_attn	Add	1 × 262144 × 2688
229	LayerNorm_38_2	LayerNorm	1 × 262144 × 2688
230	MoE_38	Shared-Expert MoE	1 × 262144 × 2688
231	Add_38_ffn	Add	1 × 262144 × 2688
232	LayerNorm_39_1	LayerNorm	1 × 262144 × 2688
233	Attention_39	Grouped Query Attn	1 × 262144 × 2688
234	Add_39_attn	Add	1 × 262144 × 2688
235	LayerNorm_39_2	LayerNorm	1 × 262144 × 2688
236	MoE_39	Shared-Expert MoE	1 × 262144 × 2688
237	Add_39_ffn	Add	1 × 262144 × 2688
238	LayerNorm_40_1	LayerNorm	1 × 262144 × 2688
239	Attention_40	Grouped Query Attn	1 × 262144 × 2688
240	Add_40_attn	Add	1 × 262144 × 2688
241	LayerNorm_40_2	LayerNorm	1 × 262144 × 2688
242	MoE_40	Shared-Expert MoE	1 × 262144 × 2688
243	Add_40_ffn	Add	1 × 262144 × 2688
244	LayerNorm_41_1	LayerNorm	1 × 262144 × 2688
245	Attention_41	Grouped Query Attn	1 × 262144 × 2688
246	Add_41_attn	Add	1 × 262144 × 2688
247	LayerNorm_41_2	LayerNorm	1 × 262144 × 2688
248	MoE_41	Shared-Expert MoE	1 × 262144 × 2688
249	Add_41_ffn	Add	1 × 262144 × 2688
250	LayerNorm_42_1	LayerNorm	1 × 262144 × 2688
251	Attention_42	Grouped Query Attn	1 × 262144 × 2688
252	Add_42_attn	Add	1 × 262144 × 2688
253	LayerNorm_42_2	LayerNorm	1 × 262144 × 2688
254	MoE_42	Shared-Expert MoE	1 × 262144 × 2688
255	Add_42_ffn	Add	1 × 262144 × 2688
256	LayerNorm_43_1	LayerNorm	1 × 262144 × 2688
257	Attention_43	Grouped Query Attn	1 × 262144 × 2688
258	Add_43_attn	Add	1 × 262144 × 2688
259	LayerNorm_43_2	LayerNorm	1 × 262144 × 2688
260	MoE_43	Shared-Expert MoE	1 × 262144 × 2688
261	Add_43_ffn	Add	1 × 262144 × 2688
262	LayerNorm_44_1	LayerNorm	1 × 262144 × 2688
263	Attention_44	Grouped Query Attn	1 × 262144 × 2688
264	Add_44_attn	Add	1 × 262144 × 2688
265	LayerNorm_44_2	LayerNorm	1 × 262144 × 2688
266	MoE_44	Shared-Expert MoE	1 × 262144 × 2688
267	Add_44_ffn	Add	1 × 262144 × 2688
268	LayerNorm_45_1	LayerNorm	1 × 262144 × 2688
269	Attention_45	Grouped Query Attn	1 × 262144 × 2688
270	Add_45_attn	Add	1 × 262144 × 2688
271	LayerNorm_45_2	LayerNorm	1 × 262144 × 2688
272	MoE_45	Shared-Expert MoE	1 × 262144 × 2688
273	Add_45_ffn	Add	1 × 262144 × 2688
274	LayerNorm_46_1	LayerNorm	1 × 262144 × 2688
275	Attention_46	Grouped Query Attn	1 × 262144 × 2688
276	Add_46_attn	Add	1 × 262144 × 2688
277	LayerNorm_46_2	LayerNorm	1 × 262144 × 2688
278	MoE_46	Shared-Expert MoE	1 × 262144 × 2688
279	Add_46_ffn	Add	1 × 262144 × 2688
280	LayerNorm_47_1	LayerNorm	1 × 262144 × 2688
281	Attention_47	Grouped Query Attn	1 × 262144 × 2688
282	Add_47_attn	Add	1 × 262144 × 2688
283	LayerNorm_47_2	LayerNorm	1 × 262144 × 2688
284	MoE_47	Shared-Expert MoE	1 × 262144 × 2688
285	Add_47_ffn	Add	1 × 262144 × 2688
286	LayerNorm_48_1	LayerNorm	1 × 262144 × 2688
287	Attention_48	Grouped Query Attn	1 × 262144 × 2688
288	Add_48_attn	Add	1 × 262144 × 2688
289	LayerNorm_48_2	LayerNorm	1 × 262144 × 2688
290	MoE_48	Shared-Expert MoE	1 × 262144 × 2688
291	Add_48_ffn	Add	1 × 262144 × 2688
292	LayerNorm_49_1	LayerNorm	1 × 262144 × 2688
293	Attention_49	Grouped Query Attn	1 × 262144 × 2688
294	Add_49_attn	Add	1 × 262144 × 2688
295	LayerNorm_49_2	LayerNorm	1 × 262144 × 2688
296	MoE_49	Shared-Expert MoE	1 × 262144 × 2688
297	Add_49_ffn	Add	1 × 262144 × 2688
298	LayerNorm_50_1	LayerNorm	1 × 262144 × 2688
299	Attention_50	Grouped Query Attn	1 × 262144 × 2688
300	Add_50_attn	Add	1 × 262144 × 2688
301	LayerNorm_50_2	LayerNorm	1 × 262144 × 2688
302	MoE_50	Shared-Expert MoE	1 × 262144 × 2688
303	Add_50_ffn	Add	1 × 262144 × 2688
304	LayerNorm_51_1	LayerNorm	1 × 262144 × 2688
305	Attention_51	Grouped Query Attn	1 × 262144 × 2688
306	Add_51_attn	Add	1 × 262144 × 2688
307	LayerNorm_51_2	LayerNorm	1 × 262144 × 2688
308	MoE_51	Shared-Expert MoE	1 × 262144 × 2688
309	Add_51_ffn	Add	1 × 262144 × 2688
310	LayerNorm_52_1	LayerNorm	1 × 262144 × 2688
311	Attention_52	Grouped Query Attn	1 × 262144 × 2688
312	Add_52_attn	Add	1 × 262144 × 2688
313	LayerNorm_52_2	LayerNorm	1 × 262144 × 2688
314	MoE_52	Shared-Expert MoE	1 × 262144 × 2688
315	Add_52_ffn	Add	1 × 262144 × 2688
316	Final_LayerNorm	LayerNorm	1 × 262144 × 2688
317	LM_Head	Linear	1 × 262144 × 131072
318	Output	Output	1 × 262144 × 131072

What the verifier says

infoAt 52 stacked attention layers, residual-branch outputs add up; unscaled init lets activation variance grow with depth. GPT-2/LLaMA-family models scale the residual projections by depth (N(0, 0.02 / √(2L))). Fix: Scale residual output projections by depth: nn.init.normal_(w, std=0.02 / math.sqrt(2 * n_layers))
deep-attention-default-init

Do this to your own model

Same numbers, on a model in your repo, in one command. No account.

pip install neurarch-trace
neurarch-trace nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-NVFP4 --plan --share

Paste a model instead Markdown twin Index JSON POST /api/v1/plan

Other nemotron_h checkpoints

NVIDIA-Nemotron-3-Nano-30B-A3B-BF16

102B derived · +224% against the checkpoint

NVIDIA-Nemotron-3-Nano-30B-A3B-FP8

102B derived · +224% against the checkpoint

NVIDIA-Nemotron-3-Super-120B-A12B-BF16

1496B derived · +1110% against the checkpoint

NVIDIA-Nemotron-3-Super-120B-A12B-FP8

1496B derived · +1110% against the checkpoint