mirror of
https://github.com/ollama/ollama.git
synced 2026-10-21 02:51:35 +00:00
* mlx: match publisher tokenizer semantics Honor pretokenizer stage order, split behavior, Unicode boundaries, added-token normalization, and ranked BPE merges. Handle empty added tokens and empty Metaspace input consistently. Add shared Go/Python reference cases using published tokenizers, pulling missing models directly and failing on errors, plus focused regressions for configuration precedence, byte fallback, and parallel encoding. * address comments |
||
|---|---|---|
| .. | ||
| testdata | ||
| tokenizer.go | ||
| tokenizer_added.go | ||
| tokenizer_benchmark_test.go | ||
| tokenizer_bpe.go | ||
| tokenizer_bpe_spans_test.go | ||
| tokenizer_config_test.go | ||
| tokenizer_correctness_test.go | ||
| tokenizer_decode.go | ||
| tokenizer_encode.go | ||
| tokenizer_ggml_parity_test.go | ||
| tokenizer_load.go | ||
| tokenizer_load_test.go | ||
| tokenizer_normalization_test.go | ||
| tokenizer_pretokenizer.go | ||
| tokenizer_pretokenizer_test.go | ||
| tokenizer_reference_benchmark_test.go | ||
| tokenizer_reference_model_test.go | ||
| tokenizer_reference_test.go | ||