Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
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
6 changes: 4 additions & 2 deletions common/chat.cpp
Original file line number Diff line number Diff line change
Expand Up @@ -3148,7 +3148,8 @@ static common_chat_params common_chat_params_init_muse_glimmer(const common_chat
auto analysis = p.ref("analysis");

auto recipient = p.optional(p.literal(" to=user"));
auto final_msg = p.rule("final", recipient + p.literal("<|message|>") + p.content(p.until("<|eot|>")));
auto final_msg = p.rule("final", recipient + p.literal("<|message|>") +
p.content(p.until_one_of({ "<|eot|>", "<|eom|>" })));

if (has_tools && inputs.tool_choice != COMMON_CHAT_TOOL_CHOICE_NONE) {
auto string_value = p.ac(
Expand Down Expand Up @@ -3204,7 +3205,8 @@ static common_chat_params common_chat_params_init_muse_glimmer(const common_chat
if (inputs.tool_choice == COMMON_CHAT_TOOL_CHOICE_REQUIRED) {
return p.zero_or_more(start + analysis) + start + tool_calls;
}
return p.zero_or_more(start + analysis) + start + (tool_calls | final_msg);
auto trailing_calls = p.optional(p.literal("<|eom|>") + start + tool_calls);
return p.zero_or_more(start + analysis) + start + (tool_calls | (final_msg + trailing_calls));
}

return p.zero_or_more(start + analysis) + start + final_msg;
Expand Down
211 changes: 211 additions & 0 deletions models/templates/muse-glimmer.jinja
Original file line number Diff line number Diff line change
@@ -0,0 +1,211 @@
{#
Template: Muse Glimmer ATEM Chat Template
Renders the ATEM tool-calling protocol: reasoning channel (to=self), tool
channels (to=<tool>), and the user channel, plus tool definitions and the
valid-recipient list in the system block.

Whitespace note: every tag uses the {%- -%} / {{- -}} stripping markers, so
the indentation below is purely for readability and contributes nothing to
the rendered output.
#}
{%- macro render_content(content) -%}
{%- if content is string -%}
{{- content -}}
{%- elif content is not none -%}
{%- for part in content -%}
{%- if part['type'] == 'image' -%}
{{- '<|patch|>' -}}
{%- elif part['type'] == 'video' -%}
{{- '<|video|>' -}}
{%- elif part['type'] == 'text' -%}
{{- part['text'] -}}
{%- endif -%}
{%- endfor -%}
{%- endif -%}
{%- endmacro -%}
{%- macro render_atem(tc) -%}
{%- set args = tc.function.arguments -%}
{%- if args is not mapping -%}
{{- raise_exception('Muse Glimmer ATEM chat template requires tool_call.function.arguments to be a dict (mapping); a JSON string cannot be parsed in the HF jinja sandbox.') -}}
{%- endif -%}
{{- '<atem:function_calls>\n<atem:invoke name="' + tc.function.name + '">\n' -}}
{%- for k, v in args.items() -%}
{{- '<atem:parameter name="' + k + '">' -}}
{%- if v is boolean -%}
{%- if v -%}
true
{%- else -%}
false
{%- endif -%}
{%- elif v is none -%}
null
{%- elif v is mapping or (v is iterable and v is not string) -%}
{{- v | tojson -}}
{%- else -%}
{{- v -}}
{%- endif -%}
{{- '</atem:parameter>\n' -}}
{%- endfor -%}
{{- '</atem:invoke>\n</atem:function_calls>' -}}
{%- endmacro -%}
{%- macro render_tool_defs(tools) -%}
{{- 'In this environment you have access to a set of tools you can use to answer the user\'s question.\n\n' -}}
{{- 'You can invoke a function by writing a "<atem:function_calls>" block like the following:\n' -}}
{{- '<atem:function_calls>\n<atem:invoke name="$FUNCTION_NAME">\n<atem:parameter name="$PARAMETER_NAME">$PARAMETER_VALUE</atem:parameter>\n...\n</atem:invoke>\n</atem:function_calls>\n\n' -}}
{{- 'String and scalar parameters should be specified as is, while lists and objects should use JSON format. Note that spaces for string values are not stripped. The output is not expected to be valid XML and is parsed with regular expressions.\n' -}}
{{- 'Here are the functions available in JSONSchema format:\n' -}}
{{- '// Tool metadata\n' -}}
{%- set nsns = namespace(seen=[]) -%}
{%- for tool in tools -%}
{%- set fn = tool.function if tool.function is defined else tool -%}
{%- set tns = fn.name.split('.')[0] -%}
{%- if tns not in nsns.seen -%}
{%- set nsns.seen = nsns.seen + [tns] -%}
{%- endif -%}
{%- endfor -%}
{%- set nd = tool_namespace_descriptions if tool_namespace_descriptions is defined else {} -%}
{%- for tns in nsns.seen -%}
{{- '{"name": ' + (tns | tojson) + ', "description": ' + ((nd[tns] if tns in nd else '') | tojson) + '}\n' -}}
{%- endfor -%}
{{- '// Function schemas' -}}
{%- for tool in tools -%}
{%- set fn = tool.function if tool.function is defined else tool -%}
{{- '\n{"name": ' + (fn.name | tojson) + ', "description": ' + (fn.description | tojson) + ', "parameters": ' + (fn.parameters | tojson) + '}' -}}
{%- endfor -%}
{{- '\n\nHere\'s an example of how to call a function in the tool set:\n' -}}
{{- '(If the tool namespace is not specified, invoke the function directly as `example_function_name` rather than `example_tool_name.example_function_name`)\n\n' -}}
{{- 'to=example_tool_name.example_function_name\n\n' -}}
{{- '<atem:function_calls>\n<atem:invoke name="example_tool_name.example_function_name">\n' -}}
{{- '<atem:parameter name="example_parameter_1">value_1</atem:parameter>\n' -}}
{{- '<atem:parameter name="example_parameter_2">This is the value for the second parameter\nthat can span\n"multiple" lines\n</atem:parameter>\n' -}}
{{- '</atem:invoke>\n</atem:function_calls>' -}}
{%- endmacro -%}
{%- macro render_reasoning() -%}
{%- set rs = reasoning_strength if reasoning_strength is defined and reasoning_strength else 'high' -%}

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

OAI uses reasoning_effort - wondering if that would be a better name? also for consistency with other templates e.g. deepseek and gpt-oss-120b

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The model has been trained explicitly with "reasoning strength", not "reasoning effort". If people actually say reasoning effort the behavior is undefined, I wanted to keep as aligned with training as possible

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I don't see how it would be harmful to set rs to reasoning_strength if that is defined, and if not, then set it to reasoning_effort if that is defined. It would certainly improve the consistency and ease of use with other models, and the model never sees the name of the chat template variables.

The real problem (in my opinion) is that llama.cpp still has no native concept of reasoning levels or how to map them to what the model is expecting. I don't think users should have to control this with a raw chat template variable. The usability problem is getting worse all the time as more models release with various reasoning levels.

But, my opinion on this stuff doesn't really matter.

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@coder543 this is my idea:

sobakasu@29d426a

the jinja template for each model would be responsible for mapping OAI resoning_effort to the model's accepted values / equivalent concept.

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

the jinja template for each model would be responsible for mapping OAI resoning_effort to the model's accepted values / equivalent concept

Just to confirm, you mean the jinja template that lives in HuggingFace, right? That is usually a standalone model artifact, that is not really aware of llama.cpp innerworks at all.

I think expecting model templates to even be aware of some OpenAI flag compatibility doesn't sound great. Isn't it better if this translation layer lives in llama.cpp directly? If llama.cpp wants a unified "reasoning_effort" flag that's fine, but then there's some code that maps that to each individual template (so in my case reasoning_effort->reasoning_strength) internally before reaching jinja.

Otherwise you would be forcing jinja files in HF to somehow have to work around serving logic that is not model specific

Let me know if Im missing anything

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I thought we were talking about a jinja file that lives in this repo, not huggingface? And it's not really OpenAI compatibility. Other models like DeepSeek V4, Kimi K3, and Hy3 use the same config variable name of reasoning_effort. reasoning_effort has been the standard variable name for models with named reasoning levels across all open weight models in the industry that I'm aware of until this model introduced reasoning_strength.

I see it as the same thing as standard role names like assistant and user. There's technically nothing that requires models to use those in the jinja template, but breaking compatibility there would make integration with standard inference runtimes harder. Unless there is a good reason for changing reasoning_effort to reasoning_strength, then I wish companies wouldn't. It only makes things harder for people looking to integrate with this model. There are surely tons of clients that already hard code reasoning_effort which will have to be changed unless llama.cpp (and vLLM, and SGLang, and every other runtime) adds a compatibility shim just for this model.

But again, my opinion doesn't matter here. I'm just sharing some of my perspective as someone who has closely tracked open weight models for years. The addition of reasoning levels in this model is awesome, and I appreciate that, although I haven't found any benchmarks that show how much the reasoning levels help.

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The template should be the one from the HF repository. The only exceptions are models that don't ship with a template.

{{- 'Reasoning strength: ' + rs + '.' -}}
{%- endmacro -%}
{%- macro render_system_meta(tools) -%}
{%- set rns = namespace(recipients=['"self"'], nslist=[]) -%}
{%- if tools -%}
{%- for tool in tools -%}
{%- set fn = tool.function if tool.function is defined else tool -%}
{%- set tns = fn.name.split('.')[0] -%}
{%- if tns not in rns.nslist -%}
{%- set rns.nslist = rns.nslist + [tns] -%}
{%- endif -%}
{%- endfor -%}
{%- for tns in rns.nslist -%}
{%- set rns.recipients = rns.recipients + ['"' + tns + '.*"'] -%}
{%- endfor -%}
{%- endif -%}
{%- set rns.recipients = rns.recipients + ['"user"'] -%}
{{- '# Valid recipients: ' + rns.recipients | join(', ') + '.' -}}
{%- endmacro -%}
{{- bos_token -}}
{%- set ns = namespace(has_system=false) -%}
{%- for m in messages -%}
{%- if m['role'] == 'system' -%}
{%- set ns.has_system = true -%}
{%- endif -%}
{%- endfor -%}
{%- if not ns.has_system -%}
{{- '<|start|>system<|message|>You are a helpful AI assistant.' -}}
{%- set kc = knowledge_cutoff if knowledge_cutoff is defined and knowledge_cutoff else '2026-01-04' -%}
{{- '\nKnowledge cutoff: ' + kc + '.' -}}
{%- if current_date is defined and current_date -%}
{{- '\nCurrent date: ' + current_date + '.' -}}
{%- elif strftime_now is defined -%}
{{- '\nCurrent date: ' + strftime_now('%Y-%m-%d') + '.' -}}
{%- endif -%}
{{- '\n\n' -}}
{{- render_reasoning() -}}
{%- if tools -%}
{{- '\n\n' -}}
{{- render_tool_defs(tools) -}}
{%- endif -%}
{{- '\n\n' -}}
{{- render_system_meta(tools) -}}
{{- '<|eot|>' -}}
{%- endif -%}
{%- for message in messages -%}
{%- set role = message['role'] -%}
{%- set end_token = '<|eom|>' if (not loop.last and messages[loop.index0 + 1]['role'] == role) else '<|eot|>' -%}
{%- if role == 'system' -%}
{#- Callers sometimes write the directive into the system prompt themselves.
Normalise "Reasoning effort" to "Reasoning strength" (jinja has no
case-insensitive replace, hence the four realistic casings), then skip
the kwarg-driven line below if the prompt already carries one. -#}
{%- set sys_text = render_content(message['content'])
| replace('Reasoning effort', 'Reasoning strength')
| replace('Reasoning Effort', 'Reasoning Strength')
| replace('reasoning effort', 'reasoning strength')
| replace('REASONING EFFORT', 'REASONING STRENGTH') -%}
{{- '<|start|>system<|message|>' -}}
{{- sys_text -}}
{%- if 'reasoning strength' not in (sys_text | lower) -%}
{{- '\n\n' -}}
{{- render_reasoning() -}}
{%- endif -%}
{%- if tools -%}
{{- '\n\n' -}}
{{- render_tool_defs(tools) -}}
{%- endif -%}
{{- '\n\n' -}}
{{- render_system_meta(tools) -}}
{{- '<|eot|>' -}}
{%- elif role == 'user' -%}
{{- '<|start|>user<|message|>' -}}
{{- render_content(message['content']) -}}
{{- '<|eot|>' -}}
{%- elif role == 'tool' -%}
{%- set tname = message.get('name') -%}
{%- if not tname -%}
{%- set tcid = message.get('tool_call_id') -%}
{%- set rns = namespace(name=tcid if tcid else '') -%}
{%- for m in messages -%}
{%- if m.get('tool_calls') -%}
{%- for tc in m['tool_calls'] -%}
{%- if tcid is not none and tc.id is defined and tc.id == tcid -%}
{%- set rns.name = tc.function.name -%}
{%- endif -%}
{%- endfor -%}
{%- endif -%}
{%- endfor -%}
{%- set tname = rns.name -%}
{%- endif -%}
{{- '<|start|>tool ' + tname + '<|message|><tool_output name="' + tname + '">\n' -}}
{{- render_content(message['content']) -}}
{{- '\n</tool_output><|eot|>' -}}
{%- elif role == 'assistant' -%}
{%- if message.get('reasoning_content') -%}
{{- '<|start|>assistant to=self<|message|>' + message['reasoning_content'] + '<|eom|>' -}}
{%- endif -%}
{%- if message.get('tool_calls') -%}
{%- for tc in message['tool_calls'] -%}
{{- '<|start|>assistant to=' + tc.function.name + '<|message|>' -}}
{{- render_atem(tc) -}}
{%- if loop.last -%}
{{- end_token -}}
{%- else -%}
{{- '<|eom|>' -}}
{%- endif -%}
{%- endfor -%}
{%- else -%}
{%- set recipient = message.get('recipient') or 'user' -%}
{%- set end_turn = message.get('end_turn') -%}
{%- if end_turn is none -%}
{%- set end_turn = not (recipient and recipient != 'user') -%}
{%- endif -%}
{{- '<|start|>assistant' -}}
{%- if recipient -%}
{{- ' to=' + recipient -}}
{%- endif -%}
{{- '<|message|>' -}}
{{- render_content(message['content']) -}}
{{- ('<|eot|>' if end_turn else '<|eom|>') -}}
{%- endif -%}
{%- endif -%}
{%- endfor -%}
{%- if add_generation_prompt -%}
{{- '<|start|>assistant' -}}
{%- endif -%}
46 changes: 46 additions & 0 deletions tests/test-chat.cpp
Original file line number Diff line number Diff line change
Expand Up @@ -5843,6 +5843,52 @@ static void test_template_output_peg_parsers(bool detailed_debug) {
.run();
}

// Muse Glimmer format tests
{
auto tst = peg_tester("models/templates/muse-glimmer.jinja", detailed_debug);

const std::string call_markup =
"<atem:function_calls>\n"
"<atem:invoke name=\"special_function\">\n"
"<atem:parameter name=\"arg1\">1</atem:parameter>\n"
"</atem:invoke>\n"
"</atem:function_calls>";

// A plain answer is unaffected
tst.test(" to=user<|message|>Hello, world!\nWhat's up?<|eot|>")
.reasoning_format(COMMON_REASONING_FORMAT_AUTO)
.expect(message_assist)
.run();

// "Inform then act": the model answers the user and calls a tool in ONE generation,
// closing the answer with <|eom|>. The answer must stop there rather than swallow it.
tst.test(" to=user<|message|>Hello, world!\nWhat's up?<|eom|>"
"<|start|>assistant to=special_function<|message|>" +
call_markup)
.tools({ special_function_tool })
.reasoning_format(COMMON_REASONING_FORMAT_AUTO)
.expect(message_with_content_and_tool_call("Hello, world!\nWhat's up?", "special_function",
"{\"arg1\":1}"))
.run();

// Markup quoted in an answer has no preceding <|eom|>, so it stays content instead of
// becoming an invocation the user never asked for
tst.test(" to=user<|message|>You invoke it like this:\n" + call_markup + "<|eot|>")
.tools({ special_function_tool })
.reasoning_format(COMMON_REASONING_FORMAT_AUTO)
.expect_content("You invoke it like this:\n" + call_markup)
.run();

// Tool markup inside the analysis channel is reasoning, not a call
tst.test(" to=self<|message|>I could use " + call_markup + " here<|eom|>"
"<|start|>assistant to=user<|message|>Hello!<|eot|>")
.tools({ special_function_tool })
.reasoning_format(COMMON_REASONING_FORMAT_AUTO)
.expect_reasoning("I could use " + call_markup + " here")
.expect_content("Hello!")
.run();
}

// GPT-OSS format tests
{
auto tst = peg_tester("models/templates/openai-gpt-oss-120b.jinja", detailed_debug);
Expand Down
Loading