embedding: add raw option for --embd-output-format

[SamMalayek]

This adds support for a new --embd-output-format raw option, which outputs embeddings as plain space-separated floats — without JSON formatting or embedding N: prefixes.

This is useful for downstream vector pipelines and scripting, e.g. when piping directly into NumPy or other vector processing tools.

• Existing formats (json, json+, etc.) remain unchanged.
• Default behavior is unaffected.

[SamMalayek]

@ggerganov gentle ping

[CISC]

Please update this

llama.cpp/examples/embedding/README.md

Lines 33 to 40 in 280d97b

	### --embd-output-format $'string'$
	| $'string'$ | description | |
	|------------|------------------------------|--|
	| '' | same as before | (default)
	| 'array' | single embeddings | $[[x_1,...,x_n]]$
	| | multiple embeddings | $[[x_1,...,x_n],[x_1,...,x_n],...,[x_1,...,x_n]]$
	| 'json' | openai style |
	| 'json+' | add cosine similarity matrix |

and this

llama.cpp/common/arg.cpp

Lines 3249 to 3255 in 280d97b

	add_opt(common_arg(
	{"--embd-output-format"}, "FORMAT",
	"empty = default, \"array\" = [[],[]...], \"json\" = openai style, \"json+\" = same \"json\" + cosine similarity matrix",
	[](common_params & params, const std::string & value) {
	params.embd_out = value;
	}
	).set_examples({LLAMA_EXAMPLE_EMBEDDING}));

[SamMalayek]

Please update this

llama.cpp/examples/embedding/README.md

Lines 33 to 40 in 280d97b

	### --embd-output-format $'string'$
	| $'string'$ | description | |
	|------------|------------------------------|--|
	| '' | same as before | (default)
	| 'array' | single embeddings | $[[x_1,...,x_n]]$
	| | multiple embeddings | $[[x_1,...,x_n],[x_1,...,x_n],...,[x_1,...,x_n]]$
	| 'json' | openai style |
	| 'json+' | add cosine similarity matrix |

and this

llama.cpp/common/arg.cpp

Lines 3249 to 3255 in 280d97b

	add_opt(common_arg(
	{"--embd-output-format"}, "FORMAT",
	"empty = default, \"array\" = [[],[]...], \"json\" = openai style, \"json+\" = same \"json\" + cosine similarity matrix",
	[](common_params & params, const std::string & value) {
	params.embd_out = value;
	}
	).set_examples({LLAMA_EXAMPLE_EMBEDDING}));

Updated docs. Thanks for pointing this out! I should have looked around the codebase and tooling more, but I actually use this raw flag for my project and wanted this pushed quickly. I also knew it was something this package really needed.

Disclaimer: revisions per CR over the years at Amazon: 1.2-1.3 (hundreds of CRs).

[SamMalayek]

One unrelated CI test — test_ctx_shift_disabled_short_prompt[-1-120-True] — failed with assert 248 == 120, which appears to be a nondeterministic failure in the context-shift tests (something I may look into for my second contribution to this project).
Could you please re-run the CI when convenient? Everything else appears to be passing cleanly.

[CISC]

One unrelated CI test — test_ctx_shift_disabled_short_prompt[-1-120-True] — failed with assert 248 == 120, which appears to be a nondeterministic failure in the context-shift tests (something I may look into for my second contribution to this project).

Don't bother, this seems to be some ccache issue, it has leaked from another branch. Nvm, model changed. :)

[SamMalayek]

Thanks again for the reviews. All changes non-functional, but good for consistency.

Currently, the embeddings CLI endpoint is in examples, has no automated tests, and is simply provided as-is (is somewhat sloppy). The below PR is the next phase:

#16923