Xuan Son Nguyen
|
a66e4a4f5d
|
make output a bit cleaner
|
2026-01-01 23:07:45 +01:00 |
Xuan Son Nguyen
|
b23b5e3c01
|
make testing more flexible
|
2026-01-01 23:02:30 +01:00 |
Xuan Son Nguyen
|
61c25c3fbf
|
trailing spaces
|
2026-01-01 22:48:42 +01:00 |
Xuan Son Nguyen
|
a10fbc77a3
|
no more std::regex
|
2026-01-01 22:48:17 +01:00 |
Xuan Son Nguyen
|
d34efd9626
|
rm type inference
|
2025-12-31 11:43:53 +01:00 |
Xuan Son Nguyen
|
cbb37dd4cd
|
improve function args handling
|
2025-12-31 11:29:40 +01:00 |
Xuan Son Nguyen
|
1b213ae5e7
|
add placeholder for tojson
|
2025-12-30 21:52:47 +01:00 |
Xuan Son Nguyen
|
4479c382ce
|
demo: type inferrence
|
2025-12-30 17:26:23 +01:00 |
Xuan Son Nguyen
|
9c0fa6f810
|
rm workarounds
|
2025-12-30 16:07:23 +01:00 |
Xuan Son Nguyen
|
9e9a70f72f
|
more fixes
|
2025-12-29 15:07:18 +01:00 |
Xuan Son Nguyen
|
026730e8e3
|
more fix, more tests
|
2025-12-29 12:53:31 +01:00 |
Xuan Son Nguyen
|
1cf25734a9
|
more tests
|
2025-12-29 10:53:32 +01:00 |
Xuan Son Nguyen
|
2a31c9a30c
|
a lot of fixes
|
2025-12-29 00:38:29 +01:00 |
Xuan Son Nguyen
|
1784a57e7b
|
impl global_from_json
|
2025-12-28 23:15:48 +01:00 |
Xuan Son Nguyen
|
55fe96a9df
|
add jinja-value.cpp
|
2025-12-28 22:49:31 +01:00 |
Xuan Son Nguyen
|
c7f246e7a5
|
allow func to access ctx
|
2025-12-28 22:15:10 +01:00 |
Xuan Son Nguyen
|
adad34f64d
|
add filter_statement
|
2025-12-28 22:02:22 +01:00 |
Xuan Son Nguyen
|
9a8a45ff3b
|
mostly works
|
2025-12-28 21:32:55 +01:00 |
Xuan Son Nguyen
|
45df0c91e7
|
testing more templates
|
2025-12-28 19:50:09 +01:00 |
Xuan Son Nguyen
|
db09a7468d
|
fix negate test
|
2025-12-28 19:07:01 +01:00 |
Xuan Son Nguyen
|
acb0effa25
|
allow print source on exception
|
2025-12-28 18:45:41 +01:00 |
Xuan Son Nguyen
|
64e29a5848
|
add mk_stmt
|
2025-12-28 17:48:14 +01:00 |
Xuan Son Nguyen
|
7f17608ea4
|
use shared_ptr for values
|
2025-12-28 17:46:25 +01:00 |
Xuan Son Nguyen
|
4331e9c8e9
|
keyword arguments and slicing array
|
2025-12-28 17:23:29 +01:00 |
Xuan Son Nguyen
|
45c194622e
|
support binded functions
|
2025-12-28 15:33:14 +01:00 |
Xuan Son Nguyen
|
4ca114b095
|
track input string even after transformations
|
2025-12-28 12:48:35 +01:00 |
Xuan Son Nguyen
|
81310d29c1
|
render gemma tmpl ok
|
2025-12-28 12:04:23 +01:00 |
Xuan Son Nguyen
|
10835f2720
|
eval with is_user_input
|
2025-12-27 23:25:20 +01:00 |
Xuan Son Nguyen
|
c08f4ddf01
|
use mk_val
|
2025-12-27 22:28:54 +01:00 |
Xuan Son Nguyen
|
da7bbe5813
|
wip
|
2025-12-27 22:25:19 +01:00 |
Xuan Son Nguyen
|
7ed11f78f9
|
add more builtins
|
2025-12-27 22:10:45 +01:00 |
Xuan Son Nguyen
|
15b3dbab05
|
add string builtins
|
2025-12-27 21:52:50 +01:00 |
Xuan Son Nguyen
|
5a041e65b8
|
fix map object
|
2025-12-27 20:38:06 +01:00 |
Xuan Son Nguyen
|
d8ef00e610
|
bin ops works!
|
2025-12-27 20:16:46 +01:00 |
Xuan Son Nguyen
|
8d1e9a0d12
|
shadow naming
|
2025-12-27 16:06:23 +01:00 |
Xuan Son Nguyen
|
7ad6eb39ca
|
binary_expression::execute
|
2025-12-27 16:00:07 +01:00 |
Xuan Son Nguyen
|
8cea1ed6b0
|
parser ok
|
2025-12-27 12:55:01 +01:00 |
Xuan Son Nguyen
|
7ac8e98b28
|
clean up
|
2025-12-27 12:35:19 +01:00 |
Xuan Son Nguyen
|
a6e0ae7a85
|
demo
|
2025-12-27 12:22:34 +01:00 |
Xuan Son Nguyen
|
a35fcb00b5
|
add vm types
|
2025-12-27 12:12:07 +01:00 |
Xuan Son Nguyen
|
15b7c50e95
|
lexer
|
2025-12-25 21:08:51 +01:00 |
Xuan Son Nguyen
|
8d8030142e
|
jinja vm
|
2025-12-25 00:19:23 +01:00 |
Xuan-Son Nguyen
|
4cbafad4f0
|
model: support MiMo-V2-Flash (#18328)
* mimov2: convert ok
* rename mimov2 --> mimo2
* fix conversion
* runnable not incorrect
* use sink
* add_sliding_window_pattern
* add swa and per-layer n_head_kv
* correct params
* somewhat working
* correct gating func
* nits
* mimo2: wire RMS eps + MoE bias + converter guards
* add co-author
Co-authored-by: Aaryan-Kapoor <Aaryan-Kapoor@users.noreply.github.com>
* use add_rope_freq_base_swa
---------
Co-authored-by: Aaryan Kapoor <aaryankapoor2006@gmail.com>
Co-authored-by: Aaryan-Kapoor <Aaryan-Kapoor@users.noreply.github.com>
|
2025-12-24 23:07:08 +01:00 |
Aadeshveer Singh
|
c184284230
|
fit-params : fix race condition in fit-params output (#18276)
|
2025-12-24 15:57:38 +01:00 |
Aman Gupta
|
c8a2417d7b
|
CUDA: experimental native mxfp4 support for blackwell (#17906)
* CUDA: experimental native mxfp4 support for blackwell
* optimize load_tiles
* optimize quantize_mxfp4
* cleanup
* first pass review: formatting
* use interleaved layout for mma
* mmq: add assert for size
* use __nv_fp4x4_e2m1
* use iter_k as 512, cleanup
* Use 1200 as blackwell instead of 1000
* address review comments
* mmq: fix stride
* quantize.cu: use reference impl of e8m0 scale
* address review comments
* add 120f-virtual + minor fixes
---------
Co-authored-by: Aman Gupta <aman>
|
2025-12-24 22:28:26 +08:00 |
Saba Fallah
|
54132f1b1f
|
model : support for LlamaBidirectionalModel architecture (#18220)
* model: llama-embed-nemotron
* minor: python lint
* changed arch-name
* templated llm_build_llama to be used for both llama and llama-embed arch
|
2025-12-24 14:02:36 +01:00 |
Jeff Bolz
|
2a9ea2020c
|
vulkan: fix command buffer corruption in ggml_backend_vk_event_wait (#18302)
|
2025-12-24 12:36:34 +01:00 |
Wang Weixuan
|
ce7a6dc0fc
|
CANN : refactor ACL graph cache (#17752)
Move the graph property checking code into methods of LRU cache.
Signed-off-by: Wang Weixuan <wangweixvan@gmail.com>
|
2025-12-24 17:50:24 +08:00 |
Jesse Ikonen
|
1ce0126b18
|
docs: Fix typos in SYCL documentation (#18269)
|
2025-12-24 17:19:47 +08:00 |
Ruben Ortlam
|
7f459c98e7
|
vulkan: use fewer FA rows for small cache runs (#18280)
|
2025-12-24 08:59:14 +01:00 |