char-fork/llama.cpp

git clone https://git.t4t.associates/char-fork/llama.cpp

Bartowskitests : add unit test coverage for llama_tensor_get_type (#20112)7992aa7c8

char/aliceai-t5-moe
38.3 KiB1452 linesraw
1# Model: gemma-3-4b-it
2# n_embd=2560, n_ff=10240, n_vocab=262144, n_layer=34, n_head=8, n_head_kv=4
3
4[F32] f32
5
6[F16] f16
7
8[Q4_0] q4_0
9token_embd.weight q6_K
10
11[Q4_1] q4_1
12token_embd.weight q6_K
13
14[Q8_0] q8_0
15
16[Q5_0] q5_0
17token_embd.weight q6_K
18
19[Q5_1] q5_1
20token_embd.weight q6_K
21
22[Q2_K] q2_K
23token_embd.weight q6_K
24blk.0.attn_output.weight q3_K
25blk.0.attn_v.weight q3_K
26blk.0.ffn_down.weight q3_K
27blk.1.attn_output.weight q3_K
28blk.1.attn_v.weight q3_K
29blk.1.ffn_down.weight q3_K
30blk.2.attn_output.weight q3_K
31blk.2.attn_v.weight q3_K
32blk.2.ffn_down.weight q3_K
33blk.3.attn_output.weight q3_K
34blk.3.attn_v.weight q3_K
35blk.3.ffn_down.weight q3_K
36blk.4.attn_output.weight q3_K
37blk.4.attn_v.weight q3_K
38blk.4.ffn_down.weight q3_K
39blk.5.attn_output.weight q3_K
40blk.5.attn_v.weight q3_K
41blk.5.ffn_down.weight q3_K
42blk.6.attn_output.weight q3_K
43blk.6.attn_v.weight q3_K
44blk.6.ffn_down.weight q3_K
45blk.7.attn_output.weight q3_K
46blk.7.attn_v.weight q3_K
47blk.7.ffn_down.weight q3_K
48blk.8.attn_output.weight q3_K
49blk.8.attn_v.weight q3_K
50blk.8.ffn_down.weight q3_K
51blk.9.attn_output.weight q3_K
52blk.9.attn_v.weight q3_K
53blk.9.ffn_down.weight q3_K
54blk.10.attn_output.weight q3_K
55blk.10.attn_v.weight q3_K
56blk.10.ffn_down.weight q3_K
57blk.11.attn_output.weight q3_K
58blk.11.attn_v.weight q3_K
59blk.11.ffn_down.weight q3_K
60blk.12.attn_output.weight q3_K
61blk.12.attn_v.weight q3_K
62blk.12.ffn_down.weight q3_K
63blk.13.attn_output.weight q3_K
64blk.13.attn_v.weight q3_K
65blk.13.ffn_down.weight q3_K
66blk.14.attn_output.weight q3_K
67blk.14.attn_v.weight q3_K
68blk.14.ffn_down.weight q3_K
69blk.15.attn_output.weight q3_K
70blk.15.attn_v.weight q3_K
71blk.15.ffn_down.weight q3_K
72blk.16.attn_output.weight q3_K
73blk.16.attn_v.weight q3_K
74blk.16.ffn_down.weight q3_K
75blk.17.attn_output.weight q3_K
76blk.17.attn_v.weight q3_K
77blk.17.ffn_down.weight q3_K
78blk.18.attn_output.weight q3_K
79blk.18.attn_v.weight q3_K
80blk.18.ffn_down.weight q3_K
81blk.19.attn_output.weight q3_K
82blk.19.attn_v.weight q3_K
83blk.19.ffn_down.weight q3_K
84blk.20.attn_output.weight q3_K
85blk.20.attn_v.weight q3_K
86blk.20.ffn_down.weight q3_K
87blk.21.attn_output.weight q3_K
88blk.21.attn_v.weight q3_K
89blk.21.ffn_down.weight q3_K
90blk.22.attn_output.weight q3_K
91blk.22.attn_v.weight q3_K
92blk.22.ffn_down.weight q3_K
93blk.23.attn_output.weight q3_K
94blk.23.attn_v.weight q3_K
95blk.23.ffn_down.weight q3_K
96blk.24.attn_output.weight q3_K
97blk.24.attn_v.weight q3_K
98blk.24.ffn_down.weight q3_K
99blk.25.attn_output.weight q3_K
100blk.25.attn_v.weight q3_K
101blk.25.ffn_down.weight q3_K
102blk.26.attn_output.weight q3_K
103blk.26.attn_v.weight q3_K
104blk.26.ffn_down.weight q3_K
105blk.27.attn_output.weight q3_K
106blk.27.attn_v.weight q3_K
107blk.27.ffn_down.weight q3_K
108blk.28.attn_output.weight q3_K
109blk.28.attn_v.weight q3_K
110blk.28.ffn_down.weight q3_K
111blk.29.attn_output.weight q3_K
112blk.29.attn_v.weight q3_K
113blk.29.ffn_down.weight q3_K
114blk.30.attn_output.weight q3_K
115blk.30.attn_v.weight q3_K
116blk.30.ffn_down.weight q3_K
117blk.31.attn_output.weight q3_K
118blk.31.attn_v.weight q3_K
119blk.31.ffn_down.weight q3_K
120blk.32.attn_output.weight q3_K
121blk.32.attn_v.weight q3_K
122blk.32.ffn_down.weight q3_K
123blk.33.attn_output.weight q3_K
124blk.33.attn_v.weight q3_K
125blk.33.ffn_down.weight q3_K
126
127[Q3_K_S] q3_K
128token_embd.weight q6_K
129
130[Q3_K_M] q3_K
131token_embd.weight q6_K
132blk.0.attn_output.weight q4_K
133blk.0.attn_v.weight q5_K
134blk.0.ffn_down.weight q5_K
135blk.1.attn_output.weight q4_K
136blk.1.attn_v.weight q5_K
137blk.1.ffn_down.weight q5_K
138blk.2.attn_output.weight q4_K
139blk.2.attn_v.weight q4_K
140blk.2.ffn_down.weight q4_K
141blk.3.attn_output.weight q4_K
142blk.3.attn_v.weight q4_K
143blk.3.ffn_down.weight q4_K
144blk.4.attn_output.weight q4_K
145blk.4.attn_v.weight q4_K
146blk.4.ffn_down.weight q4_K
147blk.5.attn_output.weight q4_K
148blk.5.attn_v.weight q4_K
149blk.5.ffn_down.weight q4_K
150blk.6.attn_output.weight q4_K
151blk.6.attn_v.weight q4_K
152blk.6.ffn_down.weight q4_K
153blk.7.attn_output.weight q4_K
154blk.7.attn_v.weight q4_K
155blk.7.ffn_down.weight q4_K
156blk.8.attn_output.weight q4_K
157blk.8.attn_v.weight q4_K
158blk.8.ffn_down.weight q4_K
159blk.9.attn_output.weight q4_K
160blk.9.attn_v.weight q4_K
161blk.9.ffn_down.weight q4_K
162blk.10.attn_output.weight q4_K
163blk.10.attn_v.weight q4_K
164blk.10.ffn_down.weight q4_K
165blk.11.attn_output.weight q4_K
166blk.11.attn_v.weight q4_K
167blk.11.ffn_down.weight q4_K
168blk.12.attn_output.weight q4_K
169blk.12.attn_v.weight q4_K
170blk.12.ffn_down.weight q4_K
171blk.13.attn_output.weight q4_K
172blk.13.attn_v.weight q4_K
173blk.13.ffn_down.weight q4_K
174blk.14.attn_output.weight q4_K
175blk.14.attn_v.weight q4_K
176blk.14.ffn_down.weight q4_K
177blk.15.attn_output.weight q4_K
178blk.15.attn_v.weight q4_K
179blk.15.ffn_down.weight q4_K
180blk.16.attn_output.weight q4_K
181blk.16.attn_v.weight q4_K
182blk.16.ffn_down.weight q4_K
183blk.17.attn_output.weight q4_K
184blk.17.attn_v.weight q4_K
185blk.17.ffn_down.weight q4_K
186blk.18.attn_output.weight q4_K
187blk.18.attn_v.weight q4_K
188blk.18.ffn_down.weight q4_K
189blk.19.attn_output.weight q4_K
190blk.19.attn_v.weight q4_K
191blk.19.ffn_down.weight q4_K
192blk.20.attn_output.weight q4_K
193blk.20.attn_v.weight q4_K
194blk.20.ffn_down.weight q4_K
195blk.21.attn_output.weight q4_K
196blk.21.attn_v.weight q4_K
197blk.21.ffn_down.weight q4_K
198blk.22.attn_output.weight q4_K
199blk.22.attn_v.weight q4_K
200blk.22.ffn_down.weight q4_K
201blk.23.attn_output.weight q4_K
202blk.23.attn_v.weight q4_K
203blk.23.ffn_down.weight q4_K
204blk.24.attn_output.weight q4_K
205blk.24.attn_v.weight q4_K
206blk.24.ffn_down.weight q4_K
207blk.25.attn_output.weight q4_K
208blk.25.attn_v.weight q4_K
209blk.25.ffn_down.weight q4_K
210blk.26.attn_output.weight q4_K
211blk.26.attn_v.weight q4_K
212blk.26.ffn_down.weight q4_K
213blk.27.attn_output.weight q4_K
214blk.27.attn_v.weight q4_K
215blk.27.ffn_down.weight q4_K
216blk.28.attn_output.weight q4_K
217blk.28.attn_v.weight q4_K
218blk.28.ffn_down.weight q4_K
219blk.29.attn_output.weight q4_K
220blk.29.attn_v.weight q4_K
221blk.29.ffn_down.weight q4_K
222blk.30.attn_output.weight q4_K
223blk.30.attn_v.weight q4_K
224blk.30.ffn_down.weight q4_K
225blk.31.attn_output.weight q4_K
226blk.31.attn_v.weight q4_K
227blk.31.ffn_down.weight q4_K
228blk.32.attn_output.weight q4_K
229blk.32.attn_v.weight q4_K
230blk.32.ffn_down.weight q4_K
231blk.33.attn_output.weight q4_K
232blk.33.attn_v.weight q4_K
233blk.33.ffn_down.weight q4_K
234
235[Q3_K_L] q3_K
236token_embd.weight q6_K
237blk.0.attn_output.weight q5_K
238blk.0.attn_v.weight q5_K
239blk.0.ffn_down.weight q5_K
240blk.1.attn_output.weight q5_K
241blk.1.attn_v.weight q5_K
242blk.1.ffn_down.weight q5_K
243blk.2.attn_output.weight q5_K
244blk.2.attn_v.weight q5_K
245blk.2.ffn_down.weight q5_K
246blk.3.attn_output.weight q5_K
247blk.3.attn_v.weight q5_K
248blk.3.ffn_down.weight q5_K
249blk.4.attn_output.weight q5_K
250blk.4.attn_v.weight q5_K
251blk.4.ffn_down.weight q5_K
252blk.5.attn_output.weight q5_K
253blk.5.attn_v.weight q5_K
254blk.5.ffn_down.weight q5_K
255blk.6.attn_output.weight q5_K
256blk.6.attn_v.weight q5_K
257blk.6.ffn_down.weight q5_K
258blk.7.attn_output.weight q5_K
259blk.7.attn_v.weight q5_K
260blk.7.ffn_down.weight q5_K
261blk.8.attn_output.weight q5_K
262blk.8.attn_v.weight q5_K
263blk.8.ffn_down.weight q5_K
264blk.9.attn_output.weight q5_K
265blk.9.attn_v.weight q5_K
266blk.9.ffn_down.weight q5_K
267blk.10.attn_output.weight q5_K
268blk.10.attn_v.weight q5_K
269blk.10.ffn_down.weight q5_K
270blk.11.attn_output.weight q5_K
271blk.11.attn_v.weight q5_K
272blk.11.ffn_down.weight q5_K
273blk.12.attn_output.weight q5_K
274blk.12.attn_v.weight q5_K
275blk.12.ffn_down.weight q5_K
276blk.13.attn_output.weight q5_K
277blk.13.attn_v.weight q5_K
278blk.13.ffn_down.weight q5_K
279blk.14.attn_output.weight q5_K
280blk.14.attn_v.weight q5_K
281blk.14.ffn_down.weight q5_K
282blk.15.attn_output.weight q5_K
283blk.15.attn_v.weight q5_K
284blk.15.ffn_down.weight q5_K
285blk.16.attn_output.weight q5_K
286blk.16.attn_v.weight q5_K
287blk.16.ffn_down.weight q5_K
288blk.17.attn_output.weight q5_K
289blk.17.attn_v.weight q5_K
290blk.17.ffn_down.weight q5_K
291blk.18.attn_output.weight q5_K
292blk.18.attn_v.weight q5_K
293blk.18.ffn_down.weight q5_K
294blk.19.attn_output.weight q5_K
295blk.19.attn_v.weight q5_K
296blk.19.ffn_down.weight q5_K
297blk.20.attn_output.weight q5_K
298blk.20.attn_v.weight q5_K
299blk.20.ffn_down.weight q5_K
300blk.21.attn_output.weight q5_K
301blk.21.attn_v.weight q5_K
302blk.21.ffn_down.weight q5_K
303blk.22.attn_output.weight q5_K
304blk.22.attn_v.weight q5_K
305blk.22.ffn_down.weight q5_K
306blk.23.attn_output.weight q5_K
307blk.23.attn_v.weight q5_K
308blk.23.ffn_down.weight q5_K
309blk.24.attn_output.weight q5_K
310blk.24.attn_v.weight q5_K
311blk.24.ffn_down.weight q5_K
312blk.25.attn_output.weight q5_K
313blk.25.attn_v.weight q5_K
314blk.25.ffn_down.weight q5_K
315blk.26.attn_output.weight q5_K
316blk.26.attn_v.weight q5_K
317blk.26.ffn_down.weight q5_K
318blk.27.attn_output.weight q5_K
319blk.27.attn_v.weight q5_K
320blk.27.ffn_down.weight q5_K
321blk.28.attn_output.weight q5_K
322blk.28.attn_v.weight q5_K
323blk.28.ffn_down.weight q5_K
324blk.29.attn_output.weight q5_K
325blk.29.attn_v.weight q5_K
326blk.29.ffn_down.weight q5_K
327blk.30.attn_output.weight q5_K
328blk.30.attn_v.weight q5_K
329blk.30.ffn_down.weight q5_K
330blk.31.attn_output.weight q5_K
331blk.31.attn_v.weight q5_K
332blk.31.ffn_down.weight q5_K
333blk.32.attn_output.weight q5_K
334blk.32.attn_v.weight q5_K
335blk.32.ffn_down.weight q5_K
336blk.33.attn_output.weight q5_K
337blk.33.attn_v.weight q5_K
338blk.33.ffn_down.weight q5_K
339
340[Q4_K_S] q4_K
341token_embd.weight q6_K
342blk.0.attn_v.weight q5_K
343blk.0.ffn_down.weight q5_K
344blk.1.attn_v.weight q5_K
345blk.1.ffn_down.weight q5_K
346blk.2.attn_v.weight q5_K
347blk.2.ffn_down.weight q5_K
348blk.3.attn_v.weight q5_K
349blk.3.ffn_down.weight q5_K
350
351[Q4_K_M] q4_K
352token_embd.weight q6_K
353blk.0.attn_v.weight q6_K
354blk.0.ffn_down.weight q6_K
355blk.1.attn_v.weight q6_K
356blk.1.ffn_down.weight q6_K
357blk.2.attn_v.weight q6_K
358blk.2.ffn_down.weight q6_K
359blk.3.attn_v.weight q6_K
360blk.3.ffn_down.weight q6_K
361blk.6.attn_v.weight q6_K
362blk.6.ffn_down.weight q6_K
363blk.9.attn_v.weight q6_K
364blk.9.ffn_down.weight q6_K
365blk.12.attn_v.weight q6_K
366blk.12.ffn_down.weight q6_K
367blk.15.attn_v.weight q6_K
368blk.15.ffn_down.weight q6_K
369blk.18.attn_v.weight q6_K
370blk.18.ffn_down.weight q6_K
371blk.21.attn_v.weight q6_K
372blk.21.ffn_down.weight q6_K
373blk.24.attn_v.weight q6_K
374blk.24.ffn_down.weight q6_K
375blk.27.attn_v.weight q6_K
376blk.27.ffn_down.weight q6_K
377blk.29.attn_v.weight q6_K
378blk.29.ffn_down.weight q6_K
379blk.30.attn_v.weight q6_K
380blk.30.ffn_down.weight q6_K
381blk.31.attn_v.weight q6_K
382blk.31.ffn_down.weight q6_K
383blk.32.attn_v.weight q6_K
384blk.32.ffn_down.weight q6_K
385blk.33.attn_v.weight q6_K
386blk.33.ffn_down.weight q6_K
387
388[Q5_K_S] q5_K
389token_embd.weight q6_K
390
391[Q5_K_M] q5_K
392token_embd.weight q6_K
393blk.0.attn_v.weight q6_K
394blk.0.ffn_down.weight q6_K
395blk.1.attn_v.weight q6_K
396blk.1.ffn_down.weight q6_K
397blk.2.attn_v.weight q6_K
398blk.2.ffn_down.weight q6_K
399blk.3.attn_v.weight q6_K
400blk.3.ffn_down.weight q6_K
401blk.6.attn_v.weight q6_K
402blk.6.ffn_down.weight q6_K
403blk.9.attn_v.weight q6_K
404blk.9.ffn_down.weight q6_K
405blk.12.attn_v.weight q6_K
406blk.12.ffn_down.weight q6_K
407blk.15.attn_v.weight q6_K
408blk.15.ffn_down.weight q6_K
409blk.18.attn_v.weight q6_K
410blk.18.ffn_down.weight q6_K
411blk.21.attn_v.weight q6_K
412blk.21.ffn_down.weight q6_K
413blk.24.attn_v.weight q6_K
414blk.24.ffn_down.weight q6_K
415blk.27.attn_v.weight q6_K
416blk.27.ffn_down.weight q6_K
417blk.29.attn_v.weight q6_K
418blk.29.ffn_down.weight q6_K
419blk.30.attn_v.weight q6_K
420blk.30.ffn_down.weight q6_K
421blk.31.attn_v.weight q6_K
422blk.31.ffn_down.weight q6_K
423blk.32.attn_v.weight q6_K
424blk.32.ffn_down.weight q6_K
425blk.33.attn_v.weight q6_K
426blk.33.ffn_down.weight q6_K
427
428[Q6_K] q6_K
429
430[IQ2_XXS] iq2_xxs
431token_embd.weight q5_K
432blk.0.attn_v.weight q2_K
433blk.0.ffn_down.weight q2_K
434blk.1.attn_v.weight q2_K
435blk.1.ffn_down.weight q2_K
436blk.2.attn_v.weight q2_K
437blk.2.ffn_down.weight q2_K
438blk.3.attn_v.weight q2_K
439blk.3.ffn_down.weight q2_K
440blk.4.attn_v.weight q2_K
441blk.5.attn_v.weight q2_K
442blk.6.attn_v.weight q2_K
443blk.7.attn_v.weight q2_K
444blk.8.attn_v.weight q2_K
445blk.9.attn_v.weight q2_K
446blk.10.attn_v.weight q2_K
447blk.11.attn_v.weight q2_K
448blk.12.attn_v.weight q2_K
449blk.13.attn_v.weight q2_K
450blk.14.attn_v.weight q2_K
451blk.15.attn_v.weight q2_K
452blk.16.attn_v.weight q2_K
453blk.17.attn_v.weight q2_K
454blk.18.attn_v.weight q2_K
455blk.19.attn_v.weight q2_K
456blk.20.attn_v.weight q2_K
457blk.21.attn_v.weight q2_K
458blk.22.attn_v.weight q2_K
459blk.23.attn_v.weight q2_K
460blk.24.attn_v.weight q2_K
461blk.25.attn_v.weight q2_K
462blk.26.attn_v.weight q2_K
463blk.27.attn_v.weight q2_K
464blk.28.attn_v.weight q2_K
465blk.29.attn_v.weight q2_K
466blk.30.attn_v.weight q2_K
467blk.31.attn_v.weight q2_K
468blk.32.attn_v.weight q2_K
469blk.33.attn_v.weight q2_K
470
471[IQ2_XS] iq2_xs
472token_embd.weight q5_K
473blk.0.attn_v.weight q2_K
474blk.0.ffn_down.weight q2_K
475blk.1.attn_v.weight q2_K
476blk.1.ffn_down.weight q2_K
477blk.2.attn_v.weight q2_K
478blk.2.ffn_down.weight q2_K
479blk.3.attn_v.weight q2_K
480blk.3.ffn_down.weight q2_K
481blk.4.attn_v.weight q2_K
482blk.5.attn_v.weight q2_K
483blk.6.attn_v.weight q2_K
484blk.7.attn_v.weight q2_K
485blk.8.attn_v.weight q2_K
486blk.9.attn_v.weight q2_K
487blk.10.attn_v.weight q2_K
488blk.11.attn_v.weight q2_K
489blk.12.attn_v.weight q2_K
490blk.13.attn_v.weight q2_K
491blk.14.attn_v.weight q2_K
492blk.15.attn_v.weight q2_K
493blk.16.attn_v.weight q2_K
494blk.17.attn_v.weight q2_K
495blk.18.attn_v.weight q2_K
496blk.19.attn_v.weight q2_K
497blk.20.attn_v.weight q2_K
498blk.21.attn_v.weight q2_K
499blk.22.attn_v.weight q2_K
500blk.23.attn_v.weight q2_K
501blk.24.attn_v.weight q2_K
502blk.25.attn_v.weight q2_K
503blk.26.attn_v.weight q2_K
504blk.27.attn_v.weight q2_K
505blk.28.attn_v.weight q2_K
506blk.29.attn_v.weight q2_K
507blk.30.attn_v.weight q2_K
508blk.31.attn_v.weight q2_K
509blk.32.attn_v.weight q2_K
510blk.33.attn_v.weight q2_K
511
512[Q2_K_S] q2_K
513token_embd.weight q6_K
514blk.0.ffn_down.weight q4_K
515blk.1.ffn_down.weight q4_K
516blk.2.ffn_down.weight q4_K
517blk.3.ffn_down.weight q4_K
518
519[IQ3_XS] iq3_s
520token_embd.weight q6_K
521blk.0.attn_k.weight iq3_xxs
522blk.0.attn_q.weight iq3_xxs
523blk.1.attn_k.weight iq3_xxs
524blk.1.attn_q.weight iq3_xxs
525blk.2.attn_k.weight iq3_xxs
526blk.2.attn_q.weight iq3_xxs
527blk.3.attn_k.weight iq3_xxs
528blk.3.attn_q.weight iq3_xxs
529blk.4.attn_k.weight iq3_xxs
530blk.4.attn_q.weight iq3_xxs
531blk.4.ffn_gate.weight iq3_xxs
532blk.4.ffn_up.weight iq3_xxs
533blk.5.attn_k.weight iq3_xxs
534blk.5.attn_q.weight iq3_xxs
535blk.5.ffn_gate.weight iq3_xxs
536blk.5.ffn_up.weight iq3_xxs
537blk.6.attn_k.weight iq3_xxs
538blk.6.attn_q.weight iq3_xxs
539blk.6.ffn_gate.weight iq3_xxs
540blk.6.ffn_up.weight iq3_xxs
541blk.7.attn_k.weight iq3_xxs
542blk.7.attn_q.weight iq3_xxs
543blk.7.ffn_gate.weight iq3_xxs
544blk.7.ffn_up.weight iq3_xxs
545blk.8.attn_k.weight iq3_xxs
546blk.8.attn_q.weight iq3_xxs
547blk.8.ffn_gate.weight iq3_xxs
548blk.8.ffn_up.weight iq3_xxs
549blk.9.attn_k.weight iq3_xxs
550blk.9.attn_q.weight iq3_xxs
551blk.9.ffn_gate.weight iq3_xxs
552blk.9.ffn_up.weight iq3_xxs
553blk.10.attn_k.weight iq3_xxs
554blk.10.attn_q.weight iq3_xxs
555blk.10.ffn_gate.weight iq3_xxs
556blk.10.ffn_up.weight iq3_xxs
557blk.11.attn_k.weight iq3_xxs
558blk.11.attn_q.weight iq3_xxs
559blk.11.ffn_gate.weight iq3_xxs
560blk.11.ffn_up.weight iq3_xxs
561blk.12.attn_k.weight iq3_xxs
562blk.12.attn_q.weight iq3_xxs
563blk.12.ffn_gate.weight iq3_xxs
564blk.12.ffn_up.weight iq3_xxs
565blk.13.attn_k.weight iq3_xxs
566blk.13.attn_q.weight iq3_xxs
567blk.13.ffn_gate.weight iq3_xxs
568blk.13.ffn_up.weight iq3_xxs
569blk.14.attn_k.weight iq3_xxs
570blk.14.attn_q.weight iq3_xxs
571blk.14.ffn_gate.weight iq3_xxs
572blk.14.ffn_up.weight iq3_xxs
573blk.15.attn_k.weight iq3_xxs
574blk.15.attn_q.weight iq3_xxs
575blk.15.ffn_gate.weight iq3_xxs
576blk.15.ffn_up.weight iq3_xxs
577blk.16.attn_k.weight iq3_xxs
578blk.16.attn_q.weight iq3_xxs
579blk.16.ffn_gate.weight iq3_xxs
580blk.16.ffn_up.weight iq3_xxs
581blk.17.attn_k.weight iq3_xxs
582blk.17.attn_q.weight iq3_xxs
583blk.17.ffn_gate.weight iq3_xxs
584blk.17.ffn_up.weight iq3_xxs
585blk.18.attn_k.weight iq3_xxs
586blk.18.attn_q.weight iq3_xxs
587blk.18.ffn_gate.weight iq3_xxs
588blk.18.ffn_up.weight iq3_xxs
589blk.19.attn_k.weight iq3_xxs
590blk.19.attn_q.weight iq3_xxs
591blk.19.ffn_gate.weight iq3_xxs
592blk.19.ffn_up.weight iq3_xxs
593blk.20.attn_k.weight iq3_xxs
594blk.20.attn_q.weight iq3_xxs
595blk.20.ffn_gate.weight iq3_xxs
596blk.20.ffn_up.weight iq3_xxs
597blk.21.attn_k.weight iq3_xxs
598blk.21.attn_q.weight iq3_xxs
599blk.21.ffn_gate.weight iq3_xxs
600blk.21.ffn_up.weight iq3_xxs
601blk.22.attn_k.weight iq3_xxs
602blk.22.attn_q.weight iq3_xxs
603blk.22.ffn_gate.weight iq3_xxs
604blk.22.ffn_up.weight iq3_xxs
605blk.23.attn_k.weight iq3_xxs
606blk.23.attn_q.weight iq3_xxs
607blk.23.ffn_gate.weight iq3_xxs
608blk.23.ffn_up.weight iq3_xxs
609blk.24.attn_k.weight iq3_xxs
610blk.24.attn_q.weight iq3_xxs
611blk.24.ffn_gate.weight iq3_xxs
612blk.24.ffn_up.weight iq3_xxs
613blk.25.attn_k.weight iq3_xxs
614blk.25.attn_q.weight iq3_xxs
615blk.25.ffn_gate.weight iq3_xxs
616blk.25.ffn_up.weight iq3_xxs
617blk.26.attn_k.weight iq3_xxs
618blk.26.attn_q.weight iq3_xxs
619blk.26.ffn_gate.weight iq3_xxs
620blk.26.ffn_up.weight iq3_xxs
621blk.27.attn_k.weight iq3_xxs
622blk.27.attn_q.weight iq3_xxs
623blk.27.ffn_gate.weight iq3_xxs
624blk.27.ffn_up.weight iq3_xxs
625blk.28.attn_k.weight iq3_xxs
626blk.28.attn_q.weight iq3_xxs
627blk.28.ffn_gate.weight iq3_xxs
628blk.28.ffn_up.weight iq3_xxs
629blk.29.attn_k.weight iq3_xxs
630blk.29.attn_q.weight iq3_xxs
631blk.30.attn_k.weight iq3_xxs
632blk.30.attn_q.weight iq3_xxs
633blk.31.attn_k.weight iq3_xxs
634blk.31.attn_q.weight iq3_xxs
635blk.32.attn_k.weight iq3_xxs
636blk.32.attn_q.weight iq3_xxs
637blk.33.attn_k.weight iq3_xxs
638blk.33.attn_q.weight iq3_xxs
639
640[IQ3_XXS] iq3_xxs
641token_embd.weight q5_K
642blk.0.attn_k.weight iq2_s
643blk.0.attn_output.weight iq3_s
644blk.0.attn_q.weight iq2_s
645blk.0.attn_v.weight iq3_s
646blk.0.ffn_down.weight q4_K
647blk.1.attn_k.weight iq2_s
648blk.1.attn_output.weight iq3_s
649blk.1.attn_q.weight iq2_s
650blk.1.attn_v.weight iq3_s
651blk.1.ffn_down.weight q4_K
652blk.2.attn_k.weight iq2_s
653blk.2.attn_output.weight iq3_s
654blk.2.attn_q.weight iq2_s
655blk.2.attn_v.weight iq3_s
656blk.2.ffn_down.weight q4_K
657blk.3.attn_k.weight iq2_s
658blk.3.attn_output.weight iq3_s
659blk.3.attn_q.weight iq2_s
660blk.3.attn_v.weight iq3_s
661blk.3.ffn_down.weight q4_K
662blk.4.attn_k.weight iq2_s
663blk.4.attn_output.weight iq3_s
664blk.4.attn_q.weight iq2_s
665blk.4.attn_v.weight iq3_s
666blk.4.ffn_down.weight q3_K
667blk.5.attn_k.weight iq2_s
668blk.5.attn_output.weight iq3_s
669blk.5.attn_q.weight iq2_s
670blk.5.attn_v.weight iq3_s
671blk.5.ffn_down.weight q3_K
672blk.6.attn_k.weight iq2_s
673blk.6.attn_output.weight iq3_s
674blk.6.attn_q.weight iq2_s
675blk.6.attn_v.weight iq3_s
676blk.6.ffn_down.weight q3_K
677blk.7.attn_k.weight iq2_s
678blk.7.attn_output.weight iq3_s
679blk.7.attn_q.weight iq2_s
680blk.7.attn_v.weight iq3_s
681blk.7.ffn_down.weight q3_K
682blk.8.attn_k.weight iq2_s
683blk.8.attn_output.weight iq3_s
684blk.8.attn_q.weight iq2_s
685blk.8.attn_v.weight iq3_s
686blk.8.ffn_down.weight q3_K
687blk.9.attn_k.weight iq2_s
688blk.9.attn_output.weight iq3_s
689blk.9.attn_q.weight iq2_s
690blk.9.attn_v.weight iq3_s
691blk.9.ffn_down.weight q3_K
692blk.10.attn_k.weight iq2_s
693blk.10.attn_output.weight iq3_s
694blk.10.attn_q.weight iq2_s
695blk.10.attn_v.weight iq3_s
696blk.10.ffn_down.weight q3_K
697blk.11.attn_k.weight iq2_s
698blk.11.attn_output.weight iq3_s
699blk.11.attn_q.weight iq2_s
700blk.11.attn_v.weight iq3_s
701blk.11.ffn_down.weight q3_K
702blk.12.attn_k.weight iq2_s
703blk.12.attn_output.weight iq3_s
704blk.12.attn_q.weight iq2_s
705blk.12.attn_v.weight iq3_s
706blk.12.ffn_down.weight q3_K
707blk.13.attn_k.weight iq2_s
708blk.13.attn_output.weight iq3_s
709blk.13.attn_q.weight iq2_s
710blk.13.attn_v.weight iq3_s
711blk.13.ffn_down.weight q3_K
712blk.14.attn_k.weight iq2_s
713blk.14.attn_output.weight iq3_s
714blk.14.attn_q.weight iq2_s
715blk.14.attn_v.weight iq3_s
716blk.14.ffn_down.weight q3_K
717blk.15.attn_k.weight iq2_s
718blk.15.attn_output.weight iq3_s
719blk.15.attn_q.weight iq2_s
720blk.15.attn_v.weight iq3_s
721blk.15.ffn_down.weight q3_K
722blk.16.attn_k.weight iq2_s
723blk.16.attn_output.weight iq3_s
724blk.16.attn_q.weight iq2_s
725blk.16.attn_v.weight iq3_s
726blk.16.ffn_down.weight q3_K
727blk.17.attn_k.weight iq2_s
728blk.17.attn_output.weight iq3_s
729blk.17.attn_q.weight iq2_s
730blk.17.attn_v.weight iq3_s
731blk.17.ffn_down.weight q3_K
732blk.18.attn_k.weight iq2_s
733blk.18.attn_output.weight iq3_s
734blk.18.attn_q.weight iq2_s
735blk.18.attn_v.weight iq3_s
736blk.18.ffn_down.weight q3_K
737blk.19.attn_k.weight iq2_s
738blk.19.attn_output.weight iq3_s
739blk.19.attn_q.weight iq2_s
740blk.19.attn_v.weight iq3_s
741blk.19.ffn_down.weight q3_K
742blk.20.attn_k.weight iq2_s
743blk.20.attn_output.weight iq3_s
744blk.20.attn_q.weight iq2_s
745blk.20.attn_v.weight iq3_s
746blk.20.ffn_down.weight q3_K
747blk.21.attn_k.weight iq2_s
748blk.21.attn_output.weight iq3_s
749blk.21.attn_q.weight iq2_s
750blk.21.attn_v.weight iq3_s
751blk.21.ffn_down.weight q3_K
752blk.22.attn_k.weight iq2_s
753blk.22.attn_output.weight iq3_s
754blk.22.attn_q.weight iq2_s
755blk.22.attn_v.weight iq3_s
756blk.22.ffn_down.weight q3_K
757blk.23.attn_k.weight iq2_s
758blk.23.attn_output.weight iq3_s
759blk.23.attn_q.weight iq2_s
760blk.23.attn_v.weight iq3_s
761blk.23.ffn_down.weight q3_K
762blk.24.attn_k.weight iq2_s
763blk.24.attn_output.weight iq3_s
764blk.24.attn_q.weight iq2_s
765blk.24.attn_v.weight iq3_s
766blk.24.ffn_down.weight q3_K
767blk.25.attn_k.weight iq2_s
768blk.25.attn_output.weight iq3_s
769blk.25.attn_q.weight iq2_s
770blk.25.attn_v.weight iq3_s
771blk.25.ffn_down.weight q3_K
772blk.26.attn_k.weight iq2_s
773blk.26.attn_output.weight iq3_s
774blk.26.attn_q.weight iq2_s
775blk.26.attn_v.weight iq3_s
776blk.26.ffn_down.weight q3_K
777blk.27.attn_k.weight iq2_s
778blk.27.attn_output.weight iq3_s
779blk.27.attn_q.weight iq2_s
780blk.27.attn_v.weight iq3_s
781blk.27.ffn_down.weight q3_K
782blk.28.attn_k.weight iq2_s
783blk.28.attn_output.weight iq3_s
784blk.28.attn_q.weight iq2_s
785blk.28.attn_v.weight iq3_s
786blk.28.ffn_down.weight q3_K
787blk.29.attn_k.weight iq2_s
788blk.29.attn_output.weight iq3_s
789blk.29.attn_q.weight iq2_s
790blk.29.attn_v.weight iq3_s
791blk.29.ffn_down.weight q3_K
792blk.30.attn_k.weight iq2_s
793blk.30.attn_output.weight iq3_s
794blk.30.attn_q.weight iq2_s
795blk.30.attn_v.weight iq3_s
796blk.30.ffn_down.weight q3_K
797blk.31.attn_k.weight iq2_s
798blk.31.attn_output.weight iq3_s
799blk.31.attn_q.weight iq2_s
800blk.31.attn_v.weight iq3_s
801blk.31.ffn_down.weight q3_K
802blk.32.attn_k.weight iq2_s
803blk.32.attn_output.weight iq3_s
804blk.32.attn_q.weight iq2_s
805blk.32.attn_v.weight iq3_s
806blk.32.ffn_down.weight q3_K
807blk.33.attn_k.weight iq2_s
808blk.33.attn_output.weight iq3_s
809blk.33.attn_q.weight iq2_s
810blk.33.attn_v.weight iq3_s
811blk.33.ffn_down.weight q3_K
812
813[IQ1_S] iq1_s
814token_embd.weight q5_K
815blk.0.attn_output.weight iq2_xxs
816blk.0.attn_v.weight q2_K
817blk.0.ffn_down.weight q2_K
818blk.1.attn_output.weight iq2_xxs
819blk.1.attn_v.weight q2_K
820blk.1.ffn_down.weight q2_K
821blk.2.attn_output.weight iq2_xxs
822blk.2.attn_v.weight q2_K
823blk.2.ffn_down.weight q2_K
824blk.3.attn_output.weight iq2_xxs
825blk.3.attn_v.weight q2_K
826blk.3.ffn_down.weight q2_K
827blk.4.attn_output.weight iq2_xxs
828blk.4.attn_v.weight q2_K
829blk.5.attn_output.weight iq2_xxs
830blk.5.attn_v.weight q2_K
831blk.6.attn_output.weight iq2_xxs
832blk.6.attn_v.weight q2_K
833blk.7.attn_output.weight iq2_xxs
834blk.7.attn_v.weight q2_K
835blk.8.attn_output.weight iq2_xxs
836blk.8.attn_v.weight q2_K
837blk.9.attn_output.weight iq2_xxs
838blk.9.attn_v.weight q2_K
839blk.10.attn_output.weight iq2_xxs
840blk.10.attn_v.weight q2_K
841blk.11.attn_output.weight iq2_xxs
842blk.11.attn_v.weight q2_K
843blk.12.attn_output.weight iq2_xxs
844blk.12.attn_v.weight q2_K
845blk.13.attn_output.weight iq2_xxs
846blk.13.attn_v.weight q2_K
847blk.14.attn_output.weight iq2_xxs
848blk.14.attn_v.weight q2_K
849blk.15.attn_output.weight iq2_xxs
850blk.15.attn_v.weight q2_K
851blk.16.attn_output.weight iq2_xxs
852blk.16.attn_v.weight q2_K
853blk.17.attn_output.weight iq2_xxs
854blk.17.attn_v.weight q2_K
855blk.18.attn_output.weight iq2_xxs
856blk.18.attn_v.weight q2_K
857blk.19.attn_output.weight iq2_xxs
858blk.19.attn_v.weight q2_K
859blk.20.attn_output.weight iq2_xxs
860blk.20.attn_v.weight q2_K
861blk.21.attn_output.weight iq2_xxs
862blk.21.attn_v.weight q2_K
863blk.22.attn_output.weight iq2_xxs
864blk.22.attn_v.weight q2_K
865blk.23.attn_output.weight iq2_xxs
866blk.23.attn_v.weight q2_K
867blk.24.attn_output.weight iq2_xxs
868blk.24.attn_v.weight q2_K
869blk.25.attn_output.weight iq2_xxs
870blk.25.attn_v.weight q2_K
871blk.26.attn_output.weight iq2_xxs
872blk.26.attn_v.weight q2_K
873blk.27.attn_output.weight iq2_xxs
874blk.27.attn_v.weight q2_K
875blk.28.attn_output.weight iq2_xxs
876blk.28.attn_v.weight q2_K
877blk.29.attn_output.weight iq2_xxs
878blk.29.attn_v.weight q2_K
879blk.30.attn_output.weight iq2_xxs
880blk.30.attn_v.weight q2_K
881blk.31.attn_output.weight iq2_xxs
882blk.31.attn_v.weight q2_K
883blk.32.attn_output.weight iq2_xxs
884blk.32.attn_v.weight q2_K
885blk.33.attn_output.weight iq2_xxs
886blk.33.attn_v.weight q2_K
887
888[IQ4_NL] iq4_nl
889token_embd.weight q6_K
890blk.0.ffn_down.weight q5_K
891blk.1.ffn_down.weight q5_K
892blk.2.ffn_down.weight q5_K
893blk.3.ffn_down.weight q5_K
894
895[IQ3_S] iq3_s
896token_embd.weight q6_K
897
898[IQ3_M] iq3_s
899token_embd.weight q6_K
900blk.0.attn_output.weight q4_K
901blk.0.attn_v.weight q4_K
902blk.0.ffn_down.weight q4_K
903blk.1.attn_output.weight q4_K
904blk.1.attn_v.weight q4_K
905blk.1.ffn_down.weight q4_K
906blk.2.attn_output.weight q4_K
907blk.2.attn_v.weight q4_K
908blk.2.ffn_down.weight q4_K
909blk.3.attn_output.weight q4_K
910blk.3.attn_v.weight q4_K
911blk.3.ffn_down.weight q4_K
912blk.4.attn_output.weight q4_K
913blk.4.attn_v.weight q4_K
914blk.5.attn_output.weight q4_K
915blk.5.attn_v.weight q4_K
916blk.6.attn_output.weight q4_K
917blk.6.attn_v.weight q4_K
918blk.7.attn_output.weight q4_K
919blk.7.attn_v.weight q4_K
920blk.8.attn_output.weight q4_K
921blk.8.attn_v.weight q4_K
922blk.9.attn_output.weight q4_K
923blk.9.attn_v.weight q4_K
924blk.10.attn_output.weight q4_K
925blk.10.attn_v.weight q4_K
926blk.11.attn_output.weight q4_K
927blk.11.attn_v.weight q4_K
928blk.12.attn_output.weight q4_K
929blk.12.attn_v.weight q4_K
930blk.13.attn_output.weight q4_K
931blk.13.attn_v.weight q4_K
932blk.14.attn_output.weight q4_K
933blk.14.attn_v.weight q4_K
934blk.15.attn_output.weight q4_K
935blk.15.attn_v.weight q4_K
936blk.16.attn_output.weight q4_K
937blk.16.attn_v.weight q4_K
938blk.17.attn_output.weight q4_K
939blk.17.attn_v.weight q4_K
940blk.18.attn_output.weight q4_K
941blk.18.attn_v.weight q4_K
942blk.19.attn_output.weight q4_K
943blk.19.attn_v.weight q4_K
944blk.20.attn_output.weight q4_K
945blk.20.attn_v.weight q4_K
946blk.21.attn_output.weight q4_K
947blk.21.attn_v.weight q4_K
948blk.22.attn_output.weight q4_K
949blk.22.attn_v.weight q4_K
950blk.23.attn_output.weight q4_K
951blk.23.attn_v.weight q4_K
952blk.24.attn_output.weight q4_K
953blk.24.attn_v.weight q4_K
954blk.25.attn_output.weight q4_K
955blk.25.attn_v.weight q4_K
956blk.26.attn_output.weight q4_K
957blk.26.attn_v.weight q4_K
958blk.27.attn_output.weight q4_K
959blk.27.attn_v.weight q4_K
960blk.28.attn_output.weight q4_K
961blk.28.attn_v.weight q4_K
962blk.29.attn_output.weight q4_K
963blk.29.attn_v.weight q4_K
964blk.30.attn_output.weight q4_K
965blk.30.attn_v.weight q4_K
966blk.31.attn_output.weight q4_K
967blk.31.attn_v.weight q4_K
968blk.32.attn_output.weight q4_K
969blk.32.attn_v.weight q4_K
970blk.33.attn_output.weight q4_K
971blk.33.attn_v.weight q4_K
972
973[IQ2_S] iq2_xs
974token_embd.weight q5_K
975blk.0.attn_output.weight iq3_s
976blk.0.attn_v.weight iq3_s
977blk.0.ffn_down.weight iq3_s
978blk.1.attn_output.weight iq3_s
979blk.1.attn_v.weight iq3_s
980blk.1.ffn_down.weight iq3_s
981blk.2.attn_output.weight iq3_s
982blk.2.attn_v.weight iq3_s
983blk.2.ffn_down.weight iq3_s
984blk.3.attn_output.weight iq3_s
985blk.3.attn_v.weight iq3_s
986blk.3.ffn_down.weight iq3_s
987blk.4.attn_output.weight iq3_s
988blk.4.attn_v.weight iq3_s
989blk.5.attn_output.weight iq3_s
990blk.5.attn_v.weight iq3_s
991blk.6.attn_output.weight iq3_s
992blk.6.attn_v.weight iq3_s
993blk.7.attn_output.weight iq3_s
994blk.7.attn_v.weight iq3_s
995blk.8.attn_output.weight iq3_s
996blk.8.attn_v.weight iq3_s
997blk.9.attn_output.weight iq3_s
998blk.9.attn_v.weight iq3_s
999blk.10.attn_output.weight iq3_s
1000blk.10.attn_v.weight iq3_s
1001blk.11.attn_output.weight iq3_s
1002blk.11.attn_v.weight iq3_s
1003blk.12.attn_output.weight iq3_s
1004blk.12.attn_v.weight iq3_s
1005blk.13.attn_output.weight iq3_s
1006blk.13.attn_v.weight iq3_s
1007blk.14.attn_output.weight iq3_s
1008blk.14.attn_v.weight iq3_s
1009blk.15.attn_output.weight iq3_s
1010blk.15.attn_v.weight iq3_s
1011blk.16.attn_output.weight iq3_s
1012blk.16.attn_v.weight iq3_s
1013blk.17.attn_output.weight iq3_s
1014blk.17.attn_v.weight iq3_s
1015blk.18.attn_output.weight iq3_s
1016blk.18.attn_v.weight iq3_s
1017blk.19.attn_output.weight iq3_s
1018blk.19.attn_v.weight iq3_s
1019blk.20.attn_output.weight iq3_s
1020blk.20.attn_v.weight iq3_s
1021blk.21.attn_output.weight iq3_s
1022blk.21.attn_v.weight iq3_s
1023blk.22.attn_output.weight iq3_s
1024blk.22.attn_v.weight iq3_s
1025blk.23.attn_output.weight iq3_s
1026blk.23.attn_v.weight iq3_s
1027blk.24.attn_output.weight iq3_s
1028blk.24.attn_v.weight iq3_s
1029blk.25.attn_output.weight iq3_s
1030blk.25.attn_v.weight iq3_s
1031blk.26.attn_output.weight iq3_s
1032blk.26.attn_v.weight iq3_s
1033blk.27.attn_output.weight iq3_s
1034blk.27.attn_v.weight iq3_s
1035blk.28.attn_output.weight iq3_s
1036blk.28.attn_v.weight iq3_s
1037blk.29.attn_output.weight iq3_s
1038blk.29.attn_v.weight iq3_s
1039blk.30.attn_output.weight iq3_s
1040blk.30.attn_v.weight iq3_s
1041blk.31.attn_output.weight iq3_s
1042blk.31.attn_v.weight iq3_s
1043blk.32.attn_output.weight iq3_s
1044blk.32.attn_v.weight iq3_s
1045blk.33.attn_output.weight iq3_s
1046blk.33.attn_v.weight iq3_s
1047
1048[IQ2_M] iq2_s
1049token_embd.weight q5_K
1050blk.0.attn_output.weight iq3_s
1051blk.0.attn_v.weight iq3_s
1052blk.0.ffn_down.weight iq3_s
1053blk.1.attn_output.weight iq3_s
1054blk.1.attn_v.weight iq3_s
1055blk.1.ffn_down.weight iq3_s
1056blk.2.attn_output.weight iq3_s
1057blk.2.attn_v.weight iq3_s
1058blk.2.ffn_down.weight iq3_s
1059blk.3.attn_output.weight iq3_s
1060blk.3.attn_v.weight iq3_s
1061blk.3.ffn_down.weight iq3_s
1062blk.4.attn_output.weight iq3_s
1063blk.4.attn_v.weight iq3_s
1064blk.5.attn_output.weight iq3_s
1065blk.5.attn_v.weight iq3_s
1066blk.6.attn_output.weight iq3_s
1067blk.6.attn_v.weight iq3_s
1068blk.7.attn_output.weight iq3_s
1069blk.7.attn_v.weight iq3_s
1070blk.8.attn_output.weight iq3_s
1071blk.8.attn_v.weight iq3_s
1072blk.9.attn_output.weight iq3_s
1073blk.9.attn_v.weight iq3_s
1074blk.10.attn_output.weight iq3_s
1075blk.10.attn_v.weight iq3_s
1076blk.11.attn_output.weight iq3_s
1077blk.11.attn_v.weight iq3_s
1078blk.12.attn_output.weight iq3_s
1079blk.12.attn_v.weight iq3_s
1080blk.13.attn_output.weight iq3_s
1081blk.13.attn_v.weight iq3_s
1082blk.14.attn_output.weight iq3_s
1083blk.14.attn_v.weight iq3_s
1084blk.15.attn_output.weight iq3_s
1085blk.15.attn_v.weight iq3_s
1086blk.16.attn_output.weight iq3_s
1087blk.16.attn_v.weight iq3_s
1088blk.17.attn_output.weight iq3_s
1089blk.17.attn_v.weight iq3_s
1090blk.18.attn_output.weight iq3_s
1091blk.18.attn_v.weight iq3_s
1092blk.19.attn_output.weight iq3_s
1093blk.19.attn_v.weight iq3_s
1094blk.20.attn_output.weight iq3_s
1095blk.20.attn_v.weight iq3_s
1096blk.21.attn_output.weight iq3_s
1097blk.21.attn_v.weight iq3_s
1098blk.22.attn_output.weight iq3_s
1099blk.22.attn_v.weight iq3_s
1100blk.23.attn_output.weight iq3_s
1101blk.23.attn_v.weight iq3_s
1102blk.24.attn_output.weight iq3_s
1103blk.24.attn_v.weight iq3_s
1104blk.25.attn_output.weight iq3_s
1105blk.25.attn_v.weight iq3_s
1106blk.26.attn_output.weight iq3_s
1107blk.26.attn_v.weight iq3_s
1108blk.27.attn_output.weight iq3_s
1109blk.27.attn_v.weight iq3_s
1110blk.28.attn_output.weight iq3_s
1111blk.28.attn_v.weight iq3_s
1112blk.29.attn_output.weight iq3_s
1113blk.29.attn_v.weight iq3_s
1114blk.30.attn_output.weight iq3_s
1115blk.30.attn_v.weight iq3_s
1116blk.31.attn_output.weight iq3_s
1117blk.31.attn_v.weight iq3_s
1118blk.32.attn_output.weight iq3_s
1119blk.32.attn_v.weight iq3_s
1120blk.33.attn_output.weight iq3_s
1121blk.33.attn_v.weight iq3_s
1122
1123[IQ4_XS] iq4_xs
1124token_embd.weight q6_K
1125blk.0.ffn_down.weight q5_K
1126blk.1.ffn_down.weight q5_K
1127blk.2.ffn_down.weight q5_K
1128blk.3.ffn_down.weight q5_K
1129
1130[IQ1_M] iq1_m
1131token_embd.weight q5_K
1132blk.0.attn_output.weight iq2_xxs
1133blk.0.attn_v.weight q2_K
1134blk.0.ffn_down.weight q2_K
1135blk.1.attn_output.weight iq2_xxs
1136blk.1.attn_v.weight q2_K
1137blk.1.ffn_down.weight q2_K
1138blk.2.attn_output.weight iq2_xxs
1139blk.2.attn_v.weight q2_K
1140blk.2.ffn_down.weight q2_K
1141blk.3.attn_output.weight iq2_xxs
1142blk.3.attn_v.weight q2_K
1143blk.3.ffn_down.weight q2_K
1144blk.4.attn_output.weight iq2_xxs
1145blk.4.attn_v.weight q2_K
1146blk.5.attn_output.weight iq2_xxs
1147blk.5.attn_v.weight q2_K
1148blk.6.attn_output.weight iq2_xxs
1149blk.6.attn_v.weight q2_K
1150blk.7.attn_output.weight iq2_xxs
1151blk.7.attn_v.weight q2_K
1152blk.8.attn_output.weight iq2_xxs
1153blk.8.attn_v.weight q2_K
1154blk.9.attn_output.weight iq2_xxs
1155blk.9.attn_v.weight q2_K
1156blk.10.attn_output.weight iq2_xxs
1157blk.10.attn_v.weight q2_K
1158blk.11.attn_output.weight iq2_xxs
1159blk.11.attn_v.weight q2_K
1160blk.12.attn_output.weight iq2_xxs
1161blk.12.attn_v.weight q2_K
1162blk.13.attn_output.weight iq2_xxs
1163blk.13.attn_v.weight q2_K
1164blk.14.attn_output.weight iq2_xxs
1165blk.14.attn_v.weight q2_K
1166blk.15.attn_output.weight iq2_xxs
1167blk.15.attn_v.weight q2_K
1168blk.16.attn_output.weight iq2_xxs
1169blk.16.attn_v.weight q2_K
1170blk.17.attn_output.weight iq2_xxs
1171blk.17.attn_v.weight q2_K
1172blk.18.attn_output.weight iq2_xxs
1173blk.18.attn_v.weight q2_K
1174blk.19.attn_output.weight iq2_xxs
1175blk.19.attn_v.weight q2_K
1176blk.20.attn_output.weight iq2_xxs
1177blk.20.attn_v.weight q2_K
1178blk.21.attn_output.weight iq2_xxs
1179blk.21.attn_v.weight q2_K
1180blk.22.attn_output.weight iq2_xxs
1181blk.22.attn_v.weight q2_K
1182blk.23.attn_output.weight iq2_xxs
1183blk.23.attn_v.weight q2_K
1184blk.24.attn_output.weight iq2_xxs
1185blk.24.attn_v.weight q2_K
1186blk.25.attn_output.weight iq2_xxs
1187blk.25.attn_v.weight q2_K
1188blk.26.attn_output.weight iq2_xxs
1189blk.26.attn_v.weight q2_K
1190blk.27.attn_output.weight iq2_xxs
1191blk.27.attn_v.weight q2_K
1192blk.28.attn_output.weight iq2_xxs
1193blk.28.attn_v.weight q2_K
1194blk.29.attn_output.weight iq2_xxs
1195blk.29.attn_v.weight q2_K
1196blk.30.attn_output.weight iq2_xxs
1197blk.30.attn_v.weight q2_K
1198blk.31.attn_output.weight iq2_xxs
1199blk.31.attn_v.weight q2_K
1200blk.32.attn_output.weight iq2_xxs
1201blk.32.attn_v.weight q2_K
1202blk.33.attn_output.weight iq2_xxs
1203blk.33.attn_v.weight q2_K
1204
1205[BF16] bf16
1206
1207[TQ1_0] tq1_0
1208token_embd.weight q6_K
1209
1210[TQ2_0] tq2_0
1211token_embd.weight q6_K
1212
1213[MXFP4_MOE] mxfp4
1214token_embd.weight q8_0
1215blk.0.attn_k.weight q8_0
1216blk.0.attn_output.weight q8_0
1217blk.0.attn_q.weight q8_0
1218blk.0.attn_v.weight q8_0
1219blk.0.ffn_down.weight q8_0
1220blk.0.ffn_gate.weight q8_0
1221blk.0.ffn_up.weight q8_0
1222blk.1.attn_k.weight q8_0
1223blk.1.attn_output.weight q8_0
1224blk.1.attn_q.weight q8_0
1225blk.1.attn_v.weight q8_0
1226blk.1.ffn_down.weight q8_0
1227blk.1.ffn_gate.weight q8_0
1228blk.1.ffn_up.weight q8_0
1229blk.2.attn_k.weight q8_0
1230blk.2.attn_output.weight q8_0
1231blk.2.attn_q.weight q8_0
1232blk.2.attn_v.weight q8_0
1233blk.2.ffn_down.weight q8_0
1234blk.2.ffn_gate.weight q8_0
1235blk.2.ffn_up.weight q8_0
1236blk.3.attn_k.weight q8_0
1237blk.3.attn_output.weight q8_0
1238blk.3.attn_q.weight q8_0
1239blk.3.attn_v.weight q8_0
1240blk.3.ffn_down.weight q8_0
1241blk.3.ffn_gate.weight q8_0
1242blk.3.ffn_up.weight q8_0
1243blk.4.attn_k.weight q8_0
1244blk.4.attn_output.weight q8_0
1245blk.4.attn_q.weight q8_0
1246blk.4.attn_v.weight q8_0
1247blk.4.ffn_down.weight q8_0
1248blk.4.ffn_gate.weight q8_0
1249blk.4.ffn_up.weight q8_0
1250blk.5.attn_k.weight q8_0
1251blk.5.attn_output.weight q8_0
1252blk.5.attn_q.weight q8_0
1253blk.5.attn_v.weight q8_0
1254blk.5.ffn_down.weight q8_0
1255blk.5.ffn_gate.weight q8_0
1256blk.5.ffn_up.weight q8_0
1257blk.6.attn_k.weight q8_0
1258blk.6.attn_output.weight q8_0
1259blk.6.attn_q.weight q8_0
1260blk.6.attn_v.weight q8_0
1261blk.6.ffn_down.weight q8_0
1262blk.6.ffn_gate.weight q8_0
1263blk.6.ffn_up.weight q8_0
1264blk.7.attn_k.weight q8_0
1265blk.7.attn_output.weight q8_0
1266blk.7.attn_q.weight q8_0
1267blk.7.attn_v.weight q8_0
1268blk.7.ffn_down.weight q8_0
1269blk.7.ffn_gate.weight q8_0
1270blk.7.ffn_up.weight q8_0
1271blk.8.attn_k.weight q8_0
1272blk.8.attn_output.weight q8_0
1273blk.8.attn_q.weight q8_0
1274blk.8.attn_v.weight q8_0
1275blk.8.ffn_down.weight q8_0
1276blk.8.ffn_gate.weight q8_0
1277blk.8.ffn_up.weight q8_0
1278blk.9.attn_k.weight q8_0
1279blk.9.attn_output.weight q8_0
1280blk.9.attn_q.weight q8_0
1281blk.9.attn_v.weight q8_0
1282blk.9.ffn_down.weight q8_0
1283blk.9.ffn_gate.weight q8_0
1284blk.9.ffn_up.weight q8_0
1285blk.10.attn_k.weight q8_0
1286blk.10.attn_output.weight q8_0
1287blk.10.attn_q.weight q8_0
1288blk.10.attn_v.weight q8_0
1289blk.10.ffn_down.weight q8_0
1290blk.10.ffn_gate.weight q8_0
1291blk.10.ffn_up.weight q8_0
1292blk.11.attn_k.weight q8_0
1293blk.11.attn_output.weight q8_0
1294blk.11.attn_q.weight q8_0
1295blk.11.attn_v.weight q8_0
1296blk.11.ffn_down.weight q8_0
1297blk.11.ffn_gate.weight q8_0
1298blk.11.ffn_up.weight q8_0
1299blk.12.attn_k.weight q8_0
1300blk.12.attn_output.weight q8_0
1301blk.12.attn_q.weight q8_0
1302blk.12.attn_v.weight q8_0
1303blk.12.ffn_down.weight q8_0
1304blk.12.ffn_gate.weight q8_0
1305blk.12.ffn_up.weight q8_0
1306blk.13.attn_k.weight q8_0
1307blk.13.attn_output.weight q8_0
1308blk.13.attn_q.weight q8_0
1309blk.13.attn_v.weight q8_0
1310blk.13.ffn_down.weight q8_0
1311blk.13.ffn_gate.weight q8_0
1312blk.13.ffn_up.weight q8_0
1313blk.14.attn_k.weight q8_0
1314blk.14.attn_output.weight q8_0
1315blk.14.attn_q.weight q8_0
1316blk.14.attn_v.weight q8_0
1317blk.14.ffn_down.weight q8_0
1318blk.14.ffn_gate.weight q8_0
1319blk.14.ffn_up.weight q8_0
1320blk.15.attn_k.weight q8_0
1321blk.15.attn_output.weight q8_0
1322blk.15.attn_q.weight q8_0
1323blk.15.attn_v.weight q8_0
1324blk.15.ffn_down.weight q8_0
1325blk.15.ffn_gate.weight q8_0
1326blk.15.ffn_up.weight q8_0
1327blk.16.attn_k.weight q8_0
1328blk.16.attn_output.weight q8_0
1329blk.16.attn_q.weight q8_0
1330blk.16.attn_v.weight q8_0
1331blk.16.ffn_down.weight q8_0
1332blk.16.ffn_gate.weight q8_0
1333blk.16.ffn_up.weight q8_0
1334blk.17.attn_k.weight q8_0
1335blk.17.attn_output.weight q8_0
1336blk.17.attn_q.weight q8_0
1337blk.17.attn_v.weight q8_0
1338blk.17.ffn_down.weight q8_0
1339blk.17.ffn_gate.weight q8_0
1340blk.17.ffn_up.weight q8_0
1341blk.18.attn_k.weight q8_0
1342blk.18.attn_output.weight q8_0
1343blk.18.attn_q.weight q8_0
1344blk.18.attn_v.weight q8_0
1345blk.18.ffn_down.weight q8_0
1346blk.18.ffn_gate.weight q8_0
1347blk.18.ffn_up.weight q8_0
1348blk.19.attn_k.weight q8_0
1349blk.19.attn_output.weight q8_0
1350blk.19.attn_q.weight q8_0
1351blk.19.attn_v.weight q8_0
1352blk.19.ffn_down.weight q8_0
1353blk.19.ffn_gate.weight q8_0
1354blk.19.ffn_up.weight q8_0
1355blk.20.attn_k.weight q8_0
1356blk.20.attn_output.weight q8_0
1357blk.20.attn_q.weight q8_0
1358blk.20.attn_v.weight q8_0
1359blk.20.ffn_down.weight q8_0
1360blk.20.ffn_gate.weight q8_0
1361blk.20.ffn_up.weight q8_0
1362blk.21.attn_k.weight q8_0
1363blk.21.attn_output.weight q8_0
1364blk.21.attn_q.weight q8_0
1365blk.21.attn_v.weight q8_0
1366blk.21.ffn_down.weight q8_0
1367blk.21.ffn_gate.weight q8_0
1368blk.21.ffn_up.weight q8_0
1369blk.22.attn_k.weight q8_0
1370blk.22.attn_output.weight q8_0
1371blk.22.attn_q.weight q8_0
1372blk.22.attn_v.weight q8_0
1373blk.22.ffn_down.weight q8_0
1374blk.22.ffn_gate.weight q8_0
1375blk.22.ffn_up.weight q8_0
1376blk.23.attn_k.weight q8_0
1377blk.23.attn_output.weight q8_0
1378blk.23.attn_q.weight q8_0
1379blk.23.attn_v.weight q8_0
1380blk.23.ffn_down.weight q8_0
1381blk.23.ffn_gate.weight q8_0
1382blk.23.ffn_up.weight q8_0
1383blk.24.attn_k.weight q8_0
1384blk.24.attn_output.weight q8_0
1385blk.24.attn_q.weight q8_0
1386blk.24.attn_v.weight q8_0
1387blk.24.ffn_down.weight q8_0
1388blk.24.ffn_gate.weight q8_0
1389blk.24.ffn_up.weight q8_0
1390blk.25.attn_k.weight q8_0
1391blk.25.attn_output.weight q8_0
1392blk.25.attn_q.weight q8_0
1393blk.25.attn_v.weight q8_0
1394blk.25.ffn_down.weight q8_0
1395blk.25.ffn_gate.weight q8_0
1396blk.25.ffn_up.weight q8_0
1397blk.26.attn_k.weight q8_0
1398blk.26.attn_output.weight q8_0
1399blk.26.attn_q.weight q8_0
1400blk.26.attn_v.weight q8_0
1401blk.26.ffn_down.weight q8_0
1402blk.26.ffn_gate.weight q8_0
1403blk.26.ffn_up.weight q8_0
1404blk.27.attn_k.weight q8_0
1405blk.27.attn_output.weight q8_0
1406blk.27.attn_q.weight q8_0
1407blk.27.attn_v.weight q8_0
1408blk.27.ffn_down.weight q8_0
1409blk.27.ffn_gate.weight q8_0
1410blk.27.ffn_up.weight q8_0
1411blk.28.attn_k.weight q8_0
1412blk.28.attn_output.weight q8_0
1413blk.28.attn_q.weight q8_0
1414blk.28.attn_v.weight q8_0
1415blk.28.ffn_down.weight q8_0
1416blk.28.ffn_gate.weight q8_0
1417blk.28.ffn_up.weight q8_0
1418blk.29.attn_k.weight q8_0
1419blk.29.attn_output.weight q8_0
1420blk.29.attn_q.weight q8_0
1421blk.29.attn_v.weight q8_0
1422blk.29.ffn_down.weight q8_0
1423blk.29.ffn_gate.weight q8_0
1424blk.29.ffn_up.weight q8_0
1425blk.30.attn_k.weight q8_0
1426blk.30.attn_output.weight q8_0
1427blk.30.attn_q.weight q8_0
1428blk.30.attn_v.weight q8_0
1429blk.30.ffn_down.weight q8_0
1430blk.30.ffn_gate.weight q8_0
1431blk.30.ffn_up.weight q8_0
1432blk.31.attn_k.weight q8_0
1433blk.31.attn_output.weight q8_0
1434blk.31.attn_q.weight q8_0
1435blk.31.attn_v.weight q8_0
1436blk.31.ffn_down.weight q8_0
1437blk.31.ffn_gate.weight q8_0
1438blk.31.ffn_up.weight q8_0
1439blk.32.attn_k.weight q8_0
1440blk.32.attn_output.weight q8_0
1441blk.32.attn_q.weight q8_0
1442blk.32.attn_v.weight q8_0
1443blk.32.ffn_down.weight q8_0
1444blk.32.ffn_gate.weight q8_0
1445blk.32.ffn_up.weight q8_0
1446blk.33.attn_k.weight q8_0
1447blk.33.attn_output.weight q8_0
1448blk.33.attn_q.weight q8_0
1449blk.33.attn_v.weight q8_0
1450blk.33.ffn_down.weight q8_0
1451blk.33.ffn_gate.weight q8_0
1452blk.33.ffn_up.weight q8_0