3 people 321ab943b2 qwen model is working (#13690) ...

* qwen model is mostly working

* add Q4_K quantization support to GGUF parser, add qwen3:1.7b model

- Add Q4_K (type 12) dequantization in nn/state.py
- Add qwen3:1.7b model using Q4_K_M quantization (smaller than Q8_0)
- Make bos_token_id optional for models like Qwen3 that don't have it
- Fix line length issues and add preset parameter to SimpleTokenizer

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Opus 4.5 <[email protected]>

* smaller diff

* test dequant

* half split

* better

* simple tok

* mock token

* polish

* better

* fix

* replace

---------

Co-authored-by: Claude Opus 4.5 <[email protected]>

2025-12-15 18:00:34 -04:00

..

__init__.py

more test cleanups (#2631)

2023-12-05 16:17:57 -08:00

test_allreduce.py

minimal mstack pr to fix allreduce (#10649)

2025-06-05 15:14:53 -07:00

test_attention.py

fast tinygrad.apps.llm (#13685)

2025-12-14 21:05:21 -05:00

test_autogen.py

fix anonymous struct fields (#13610)

2025-12-07 12:56:38 -05:00

test_conv.py

hotfix: disable flaky test

2025-11-13 06:19:28 -08:00

test_device.py

device: respect compiler ContextVars (#13523)

2025-12-02 14:42:04 +03:00

test_disk_cache.py

helpers.diskcache_clear (#4436)

2024-05-05 14:19:01 -04:00

test_disk_tensor.py

feat: assert on bufferview math (#12772)

2025-10-17 14:20:08 -07:00

test_dtype_spec.py

skip test_mean_half_precision_overflow (#12331)

2025-09-29 05:15:04 -04:00

test_dtype.py

nbytes should raise an exception when size is unlimited (#11928)

2025-09-03 07:06:20 -07:00

test_elf.py

Mesa NIR backend (NAK/LLVMpipe) (#12089)

2025-10-15 17:38:33 +08:00

test_gguf.py

add Q4_K GGUF quantization support (#13700)

2025-12-15 10:17:56 -05:00

test_gradient.py

use assign in Tensor.backward (#13674)

2025-12-13 22:43:06 -05:00

test_graph_rewrite.py

add kernel spec (#12911)

2025-10-25 11:49:20 +08:00

test_hashing.py

variable bs keccak (#10731)

2025-10-23 14:10:21 -07:00

test_helpers.py

fix caching for fetch (#13544)

2025-12-03 14:34:14 +03:00

test_indexing.py

fix: add __delitem__ to Tensor with proper TypeError (#13561)

2025-12-04 00:53:08 -08:00

test_linalg.py

shared_codegen_spec and fix index spec (#12967)

2025-10-29 09:14:11 +01:00

test_linearizer_rewrite.py

early reduce simplify (#12046)

2025-09-10 21:02:46 +08:00

test_llm_server.py

qwen model is working (#13690)

2025-12-15 18:00:34 -04:00

test_llm_tokenizer.py

clean up the LLM tokenizer (#12653)

2025-10-14 14:22:01 +08:00

test_masked_st.py

move high level tests to unit (#10760)

2025-06-10 12:55:44 -04:00

test_microbenchmarks.py

improve microbenchmarks (#13492)

2025-11-29 10:15:22 -08:00

test_mnist_dataset.py

move high level tests to unit (#10760)

2025-06-10 12:55:44 -04:00

test_pattern_matcher.py

SPEC=2 is full spec, SPEC=1 is default (#12910)

2025-10-25 11:10:43 +08:00

test_rearrange_einops.py

move high level tests to unit (#10760)

2025-06-10 12:55:44 -04:00

test_rewrite_bottom_up_gate.py

Move Ops.SPECIAL arg to src (#11918)

2025-09-04 09:31:44 +02:00

test_rewrite_map.py

Revert "UOp.factor and add chain sorting (#12413)" (#12492)

2025-10-08 03:20:23 +02:00

test_schedule_cache.py

schedule: cache unbinds for consistent cache keys (#13664)

2025-12-12 17:27:42 -05:00

test_shm_tensor.py

add needs_second_gpu decorator (#13543)

2025-12-02 19:08:23 -08:00

test_simplify_valid_idx.py

add test that shows the validhack regression with bad rewrite order (#13411)

2025-11-21 13:48:30 -05:00

test_symbolic_failures.py

symbolic fuzz failure (#13367)

2025-11-19 14:21:08 -08:00

test_symbolic_shapetracker.py

delete the ShapeTracker (#12720)

2025-10-16 15:36:22 +08:00

test_tar.py

accept filename decorator [pr] (#8049)

2024-12-05 11:40:59 +08:00

test_tensor_data.py

test_tesnor_data to unit/ (#12013)

2025-09-04 19:58:27 -04:00

test_tensor_io.py

tar_extract with Tensors (#7853)

2024-12-04 17:03:19 +08:00

test_tensor_uop_representation.py

replace mop arg with vectorized index (#12695)

2025-10-15 20:50:06 +08:00

test_tinyfs.py

remove kernelize to prepare for refactors (#13463)

2025-11-26 14:18:50 -08:00

test_tqdm.py

symbolic fuzz failure (#13367)

2025-11-19 14:21:08 -08:00

test_transcendental_helpers.py

add loads at the end (#12988)

2025-10-30 10:42:19 +08:00

test_uop_resolve.py

move files into uop dir (#10399)

2025-05-18 11:38:28 -07:00

test_uop_symbolic.py

minor test_uop_symbolic updates (#13551)

2025-12-03 13:17:44 -05:00

test_uop_vmin_vmax.py

limit AND const min max to ints [pr] (#12918)

2025-10-25 16:07:52 -04:00

test_upat_compile.py

SPEC=2 is full spec, SPEC=1 is default (#12910)

2025-10-25 11:10:43 +08:00

test_viz.py

tracing: use u32 for buffer args encoding (#13472)

2025-11-28 00:19:51 +08:00

test_winograd.py

combine pm_drop_and_clauses (#12660)

2025-10-14 10:09:41 +02:00