qqc1989 commited on
Commit
26ecd57
·
verified ·
1 Parent(s): 69d1549

support ax630c

Browse files
Files changed (31) hide show
  1. qwen3-0.6b-ax630c/model.embed_tokens.weight.bfloat16.bin +3 -0
  2. qwen3-0.6b-ax630c/qwen3_p128_l0_together.axmodel +3 -0
  3. qwen3-0.6b-ax630c/qwen3_p128_l10_together.axmodel +3 -0
  4. qwen3-0.6b-ax630c/qwen3_p128_l11_together.axmodel +3 -0
  5. qwen3-0.6b-ax630c/qwen3_p128_l12_together.axmodel +3 -0
  6. qwen3-0.6b-ax630c/qwen3_p128_l13_together.axmodel +3 -0
  7. qwen3-0.6b-ax630c/qwen3_p128_l14_together.axmodel +3 -0
  8. qwen3-0.6b-ax630c/qwen3_p128_l15_together.axmodel +3 -0
  9. qwen3-0.6b-ax630c/qwen3_p128_l16_together.axmodel +3 -0
  10. qwen3-0.6b-ax630c/qwen3_p128_l17_together.axmodel +3 -0
  11. qwen3-0.6b-ax630c/qwen3_p128_l18_together.axmodel +3 -0
  12. qwen3-0.6b-ax630c/qwen3_p128_l19_together.axmodel +3 -0
  13. qwen3-0.6b-ax630c/qwen3_p128_l1_together.axmodel +3 -0
  14. qwen3-0.6b-ax630c/qwen3_p128_l20_together.axmodel +3 -0
  15. qwen3-0.6b-ax630c/qwen3_p128_l21_together.axmodel +3 -0
  16. qwen3-0.6b-ax630c/qwen3_p128_l22_together.axmodel +3 -0
  17. qwen3-0.6b-ax630c/qwen3_p128_l23_together.axmodel +3 -0
  18. qwen3-0.6b-ax630c/qwen3_p128_l24_together.axmodel +3 -0
  19. qwen3-0.6b-ax630c/qwen3_p128_l25_together.axmodel +3 -0
  20. qwen3-0.6b-ax630c/qwen3_p128_l26_together.axmodel +3 -0
  21. qwen3-0.6b-ax630c/qwen3_p128_l27_together.axmodel +3 -0
  22. qwen3-0.6b-ax630c/qwen3_p128_l2_together.axmodel +3 -0
  23. qwen3-0.6b-ax630c/qwen3_p128_l3_together.axmodel +3 -0
  24. qwen3-0.6b-ax630c/qwen3_p128_l4_together.axmodel +3 -0
  25. qwen3-0.6b-ax630c/qwen3_p128_l5_together.axmodel +3 -0
  26. qwen3-0.6b-ax630c/qwen3_p128_l6_together.axmodel +3 -0
  27. qwen3-0.6b-ax630c/qwen3_p128_l7_together.axmodel +3 -0
  28. qwen3-0.6b-ax630c/qwen3_p128_l8_together.axmodel +3 -0
  29. qwen3-0.6b-ax630c/qwen3_p128_l9_together.axmodel +3 -0
  30. qwen3-0.6b-ax630c/qwen3_post.axmodel +3 -0
  31. run_qwen3_0.6b_int8_ctx_ax630c.sh +14 -0
qwen3-0.6b-ax630c/model.embed_tokens.weight.bfloat16.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8f29acf519434862d95613b2b4f6b9d14933a5e4d16baebf8ac0b33b410acfb6
3
+ size 311164928
qwen3-0.6b-ax630c/qwen3_p128_l0_together.axmodel ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0e86f726ef8c3d18395c75e388452555de4016630748351cd7c2564b52d327f4
3
+ size 18994136
qwen3-0.6b-ax630c/qwen3_p128_l10_together.axmodel ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bec9346589dd8d47eb7f808715443935488dd12aa29c55f084746109fc998eb4
3
+ size 18994144
qwen3-0.6b-ax630c/qwen3_p128_l11_together.axmodel ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2dfed84b1d81ad3e5d461e55910790ab5106b37476cb431aa122d8abde7f8630
3
+ size 18994048
qwen3-0.6b-ax630c/qwen3_p128_l12_together.axmodel ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c68ad50e8d021411b2feb4fb6c22673193521532bbdb9a15a88cbbbe3fc96693
3
+ size 18994104
qwen3-0.6b-ax630c/qwen3_p128_l13_together.axmodel ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c13a77a87f224b7113eb9b4015428106e2fa5108270a468129d8c7b40e7aa9de
3
+ size 18994088
qwen3-0.6b-ax630c/qwen3_p128_l14_together.axmodel ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fb6503ddc35dc3363c2efe2cb0f953d2d08894939da83f789c949de3d929aa52
3
+ size 18994240
qwen3-0.6b-ax630c/qwen3_p128_l15_together.axmodel ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aced234c4ee7bc45dc3916b4076452a3b8c1da960437f04c744d57b7584bf8c5
3
+ size 18993904
qwen3-0.6b-ax630c/qwen3_p128_l16_together.axmodel ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ad6090e51d02e3a58157ecf3b0d6554cca94eceb12be4aa614b7770a7ea9b187
3
+ size 18993960
qwen3-0.6b-ax630c/qwen3_p128_l17_together.axmodel ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6ea525891f10776268ae7200b01c90a1676805af80e7fa7d18d35fe9579a878c
3
+ size 18994096
qwen3-0.6b-ax630c/qwen3_p128_l18_together.axmodel ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:99264a7a280d039b2ca94ea4855f7222df4af7c510d20bdcad588a41c4bb1189
3
+ size 18994056
qwen3-0.6b-ax630c/qwen3_p128_l19_together.axmodel ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:92469e9769f7b59142a9dd63c478cf7f6f096be4c349012c73e23ddac4773ad6
3
+ size 18994088
qwen3-0.6b-ax630c/qwen3_p128_l1_together.axmodel ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:934716e0837f7417f99f11867fef98bbcdfb13bac41c5742f45d43265c49b41e
3
+ size 18993992
qwen3-0.6b-ax630c/qwen3_p128_l20_together.axmodel ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fe1dce6bb1c4e60f609fd52834b04ff23b42a8c96fc61cb4a3b7fb211a4d35a6
3
+ size 18994200
qwen3-0.6b-ax630c/qwen3_p128_l21_together.axmodel ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:89c50942d060b92f4c5d0eb49913195a686bc5299791bb055e43b73a6849a265
3
+ size 18993984
qwen3-0.6b-ax630c/qwen3_p128_l22_together.axmodel ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b4559b5d383021d56c00a2c10de10d67722dbd964acd6964f451f9a346e915b5
3
+ size 18993888
qwen3-0.6b-ax630c/qwen3_p128_l23_together.axmodel ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c561299852caef0e91301d391c6fa8bbdb423321ac26e9b55c3b2d11e5717086
3
+ size 18994040
qwen3-0.6b-ax630c/qwen3_p128_l24_together.axmodel ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9445a71bfb25858df310322faf8d8a02fbded0a1c31738311b1f056aca662345
3
+ size 18994112
qwen3-0.6b-ax630c/qwen3_p128_l25_together.axmodel ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7aa49618ecd0721ed6b05b8d064a65f1649f27de669ab237e473a44fcedb4602
3
+ size 18993936
qwen3-0.6b-ax630c/qwen3_p128_l26_together.axmodel ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:db1f6e9c7053a74c24598a73a67409b9029a7ee68c7cff78a407dbc45f1636c9
3
+ size 18993992
qwen3-0.6b-ax630c/qwen3_p128_l27_together.axmodel ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f7a738b3f02921f9da2a1bfbceb0bc9fc51bb47fad400409dbb351946e12f3ab
3
+ size 18994152
qwen3-0.6b-ax630c/qwen3_p128_l2_together.axmodel ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b277b5e76bf17b7130aaeff7c5cc9d6d339ee0ccd25d6c7755dda660f26d91c0
3
+ size 18994064
qwen3-0.6b-ax630c/qwen3_p128_l3_together.axmodel ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:942482427efbc182418a2d54a924f1162bc54c0633c3c324c13d2b3272d94a26
3
+ size 18994072
qwen3-0.6b-ax630c/qwen3_p128_l4_together.axmodel ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b445ce04babcce0cec7d9a8b1b28733f4a0a07778b3443b45673465ed83046e5
3
+ size 18994016
qwen3-0.6b-ax630c/qwen3_p128_l5_together.axmodel ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d6e6ca864c7cfc951974e1d56e55ce612eddb683b94806b7f44b3a44d5f2d3bd
3
+ size 18994088
qwen3-0.6b-ax630c/qwen3_p128_l6_together.axmodel ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:629e08f6ca4af87c5be559f08fa15396a040e88be55337cb3e9e4e01f416cd37
3
+ size 18994104
qwen3-0.6b-ax630c/qwen3_p128_l7_together.axmodel ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fdf75db5362b88b248b54c298a5a1718939024ac08a81cd7b3559d51989d2efa
3
+ size 18994168
qwen3-0.6b-ax630c/qwen3_p128_l8_together.axmodel ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7265f56285fa808bc6492a551f138cbbd0469ce257343d399680854f98a5d84f
3
+ size 18994152
qwen3-0.6b-ax630c/qwen3_p128_l9_together.axmodel ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ecdf9d2f3938698f0599337a662b471c8606332d8b419dcfba3e3ef469aea853
3
+ size 18993944
qwen3-0.6b-ax630c/qwen3_post.axmodel ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:03178ea02f55edd76c5b48077faaac844fa88cfb6e9d6c7a53f2584868407782
3
+ size 159161836
run_qwen3_0.6b_int8_ctx_ax630c.sh ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ./main_ax650 \
2
+ --system_prompt "You are Qwen, created by Alibaba Cloud. You are a helpful assistant." \
3
+ --template_filename_axmodel "qwen3-0.6b-ax630c/qwen3_p128_l%d_together.axmodel" \
4
+ --axmodel_num 28 \
5
+ --tokenizer_type 2 \
6
+ --url_tokenizer_model "http://127.0.0.1:12345" \
7
+ --filename_post_axmodel qwen3-0.6b-ax630c/qwen3_post.axmodel \
8
+ --filename_tokens_embed qwen3-0.6b-ax630c/model.embed_tokens.weight.bfloat16.bin \
9
+ --tokens_embed_num 151936 \
10
+ --tokens_embed_size 1024 \
11
+ --use_mmap_load_embed 1 \
12
+ --live_print 1
13
+
14
+ # --kvcache_path /home/axera/ax-llm/build/kvcache_yuanqi \