Skip to content

fix: Cohere usage parsing and remove deprecated models #52

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
Kamilbenkirane merged 1 commit into from
Nov 27, 2025
Merged

fix: Cohere usage parsing and remove deprecated models #52

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 4 additions & 4 deletions packages/text-generation/src/celeste_text_generation/providers/cohere/client.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -41,10 +41,10 @@ def _init_request(self, inputs: TextGenerationInput) -> dict[str, Any]:

def _parse_usage(self, response_data: dict[str, Any]) -> TextGenerationUsage:
"""Parse usage from response."""
meta = response_data.get("meta", {})
usage_data = response_data.get("usage", {})

billed_units = meta.get("billed_units", {})
tokens = meta.get("tokens", {})
billed_units = usage_data.get("billed_units", {})
tokens = usage_data.get("tokens", {})

input_tokens = billed_units.get("input_tokens")
output_tokens = billed_units.get("output_tokens")
Expand All @@ -54,7 +54,7 @@ def _parse_usage(self, response_data: dict[str, Any]) -> TextGenerationUsage:
input_tokens=input_tokens,
output_tokens=output_tokens,
total_tokens=tokens.get("total_tokens") if tokens else None,
cached_tokens=meta.get("cached_tokens"),
cached_tokens=usage_data.get("cached_tokens"),
)

return TextGenerationUsage()
Expand Down
25 changes: 0 additions & 25 deletions packages/text-generation/src/celeste_text_generation/providers/cohere/models.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,31 +18,6 @@
# thinking_budget: Not confirmed for this model, omit constraint
},
),
Model(
id="command-r-plus",
provider=Provider.COHERE,
display_name="Command R+",
streaming=True,
parameter_constraints={
Parameter.TEMPERATURE: Range(min=0.0, max=1.0, step=0.01),
Parameter.MAX_TOKENS: Range(min=1, max=4096, step=1),
TextGenerationParameter.OUTPUT_SCHEMA: Schema(),
# command-r-plus supports reasoning (optimized for complex reasoning)
TextGenerationParameter.THINKING_BUDGET: Range(min=-1, max=31000, step=1),
},
),
Model(
id="command-r",
provider=Provider.COHERE,
display_name="Command R",
streaming=True,
parameter_constraints={
Parameter.TEMPERATURE: Range(min=0.0, max=1.0, step=0.01),
Parameter.MAX_TOKENS: Range(min=1, max=4096, step=1),
TextGenerationParameter.OUTPUT_SCHEMA: Schema(),
# thinking_budget: Support unclear, omit constraint for now
},
),
Model(
id="command-r7b-12-2024",
provider=Provider.COHERE,
Expand Down
10 changes: 5 additions & 5 deletions packages/text-generation/src/celeste_text_generation/providers/cohere/streaming.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -94,11 +94,11 @@ def _parse_chunk(self, event: dict[str, Any]) -> TextGenerationChunk | None:
else None
)

meta = event.get("meta", {})
usage_data = event.get("usage", {})
usage = None
if isinstance(meta, dict):
billed_units = meta.get("billed_units", {})
tokens = meta.get("tokens", {})
if isinstance(usage_data, dict):
billed_units = usage_data.get("billed_units", {})
tokens = usage_data.get("tokens", {})

input_tokens = billed_units.get("input_tokens")
output_tokens = billed_units.get("output_tokens")
Expand All @@ -108,7 +108,7 @@ def _parse_chunk(self, event: dict[str, Any]) -> TextGenerationChunk | None:
input_tokens=input_tokens,
output_tokens=output_tokens,
total_tokens=tokens.get("total_tokens") if tokens else None,
cached_tokens=meta.get("cached_tokens"),
cached_tokens=usage_data.get("cached_tokens"),
)

return TextGenerationChunk(
Expand Down
Loading