A drop-in replacement chat template for Qwen/Qwen3.6-27B tuned for open-source agentic coding harnesses.

A developer has released a drop-in replacement chat template for Qwen/Qwen3.6-27B that fixes six critical bugs affecting open-source agentic coding harnesses. The fork addresses issues including multi-turn tool argument collapse caused by the upstream template's `preserve_thinking=false` default, rejection of the `developer` message role used by modern coding tools, and crashes from tool call arguments arriving as JSON strings. The template also patches the upstream's failure to recognize malformed `<tool_call>` tags and wasteful token usage from passing OpenAI envelope tool definitions verbatim.

| { --------------------------------------------------------------------- | | | custom pub chat template qwen36.jinja | | | ===================================== | | | A public, harness-friendly fork of Qwen's Qwen3.6-27B chat template, | | | tuned for open-source agentic coding harnesses like: | | | - anomalyco/opencode https://github.com/anomalyco/opencode | | | - earendil-works/pi https://github.com/earendil-works/pi | | | - openclaw, OpenHarness, similar Claude-Code-style harnesses | | | WHY THIS FORK EXISTS | | | -------------------- | | | The upstream chat template at Qwen/Qwen3.6-27B is correct for chat | | | use, but six real edge cases bite agentic coding harnesses pointing | | | at a self-hosted SGLang / vLLM / llama.cpp endpoint serving Qwen3.6: | | | 1. Multi-turn tool argument collapse. After 2-3 turns of calling the | | | same tool, the model emits arguments: {} despite its prior | | | reasoning correctly identifying the parameters. Root cause: the | | | upstream template defaults preserve thinking=false, which means | | | prior-turn <think blocks are silently dropped from history; the | | | model loses its own trace of "how did I pick the parameters last | | | time?" and degenerates. Documented at: | | | https://github.com/earendil-works/pi/issues/3325 | | | The Qwen3.6 model card explicitly states the model was post- | | | trained for "Thinking Preservation" in agent scenarios — the | | | preserve thinking-FALSE default is wrong for our use case. | | | 2. The developer role rejected. Modern coding harnesses | | | opencode, Claude Code, openclaw, Continue send a developer | | | role for reasoning-capable models, following OpenAI's Responses | | | API convention. Upstream raises "Unexpected message role" — | | | crashing the entire request. Reported and documented at: | | | https://gist.github.com/sudoingX/c2facf7d8f7608c65c1024ef3b22d431 | | | "Qwen 3.5 GGUF templates reject the developer role sent by | | | OpenCode, Claude Code, and other modern agent tools." | | | 3. tool call.arguments arriving as a JSON string crashes with a | | | cryptic Jinja error "Can only get item pairs from a mapping" . | | | The Vercel AI SDK used by opencode and several other OpenAI- | | | compatible adapters hand arguments back as a JSON-encoded | | | STRING rather than the deserialized object. Diagnosing this | | | from the upstream error message is painful. Documented at: | | | https://github.com/earendil-works/pi/issues/3325 | | | https://github.com/anomalyco/opencode/issues/24264 | | | 4. The opening <tool call tag is sometimes omitted by the model | | | documented at https://github.com/QwenLM/Qwen3-Coder/issues/475 | | | and <tool call can appear inside an unclosed <think block | | | https://github.com/ollama/ollama/issues/14493 . The upstream | | | template's content-parsing only recognizes </think and only | | | when properly closed, so reasoning bleeds into the conversation | | | content channel. Whitespace variants of </think aren't | | | recognized either. | | | 5. The OpenAI envelope around tool definitions | | | {"type":"function","function":{...}} is passed verbatim | | | through tool | tojson , wasting tokens and diverging from | | | what the model expects. Qwen's own most recent coder model, | | | Qwen3-Coder-Next, unwraps this envelope in its own canonical | | | chat template: | | | https://huggingface.co/Qwen/Qwen3-Coder-Next/blob/main/chat template.jinja | | | lines 35-37 . The Qwen3.6-27B upstream template just hasn't | | | caught up to the newer convention. | | | 6. The upstream IMPORTANT instructions block is missing three | | | bullets that address the most common public Qwen3-Coder | | | failure modes: | | | - Omitting the opening <tool call tag Qwen3-Coder 475 | | | - Indenting <tool call with leading whitespace | | | https://github.com/block/goose/issues/6883 | | | - Nesting <tool call blocks instead of emitting parallel | | | calls | | | PATCH INVENTORY full details next to each patch site below | | | ------------------------------------------------------------ | | | Q1 preserve thinking default flipped FALSE→TRUE | | | Q2 developer role accepted as alias for system | | | Q3 Raise a clear, debuggable error on string tool call.arguments | | | Q4 Robust </think variant handling + unclosed-think rescue | | | Q5 Unwrap OpenAI tool envelope to inner function spec gated | | | Q6 Strengthened IMPORTANT instructions block gated | | | INVARIANTS | | | ---------- | | | 1. STRICT-EQUIVALENCE: With kwargs | | | preserve thinking=false, recovers Q1 | | | unwrap tool envelope=false, recovers Q5 | | | verbose tool instructions=false recovers Q6 | | | AND inputs that don't exercise Q2 no developer role , | | | Q3 no string-typed arguments , or Q4 no </thinking or | | | whitespace variants of </think , this template renders | | | byte-for-byte identical to upstream. The conformance suite | | | at tests/test custom pub chat template qwen36.py locks this in | | | across the simple-input matrix. | | | 2. STRICT-SAFETY: For every input upstream handles without error, | | | this template handles correctly with semantically equivalent | | | or strictly safer output. The strict-where-upstream-silent | | | patches Q3, Q4 only fire on inputs that hit the documented | | | bug surfaces. | | | USAGE | | | ----- | | | Server side e.g. SGLang or vLLM : | | | SGLang | | | python -m sglang.launch server \ | | | --model-path Qwen/Qwen3.6-27B \ | | | --chat-template /path/to/custom pub chat template qwen36.jinja \ | | | --tool-call-parser qwen3 coder \ | | | --reasoning-parser qwen3 | | | vLLM | | | vllm serve Qwen/Qwen3.6-27B \ | | | --chat-template /path/to/custom pub chat template qwen36.jinja \ | | | --tool-call-parser qwen3 coder \ | | | --reasoning-parser qwen3 \ | | | --enable-auto-tool-choice | | | Harness side: no changes required for the common case. The | | | defaults are tuned for agentic coding out of the box. If you need | | | to recover the upstream defaults explicitly: | | | { | | | "extra body": { | | | "chat template kwargs": { | | | "enable thinking": true, | | | "preserve thinking": false, | | | "unwrap tool envelope": false, | | | "verbose tool instructions": false | | | } | | | } | | | } | | | For opencode-style providers, this maps to chat template args in | | | the model config; for pi, use compat.thinkingFormat="qwen-chat- | | | template" and pi will inject the kwargs correctly. | | | PINS | | | ---- | | | Forked from Qwen/Qwen3.6-27B/chat template.jinja | | | Upstream MD5: 52b6d51ae5b203cb67e64b648494dad2 153 lines | | | Fork date: 2026-05-25 | | | License: Apache 2.0 same as upstream | | | Maintainer: see repo README | | | --------------------------------------------------------------------- } | | | { - Vision counters identical to upstream . - } | | | {%- set image count = namespace value=0 %} | | | {%- set video count = namespace value=0 %} | | | { - ============================================================================ | | | Content rendering macro. | | | Functionally identical to upstream's macro of the same name. The only | | | cosmetic difference is the add vision id is defined and add vision id | | | guard instead of upstream's bare if add vision id — a defensive | | | rewrite that prevents undefined-variable errors in some minijinja | | | runtimes llama.cpp, MLX . No rendering-time behavior change for | | | Python Jinja2 SGLang/vLLM since both runtimes treat undefined as | | | falsy. | | | ============================================================================ - } | | | {%- macro render content content, do vision count, is system content=false %} | | | {%- if content is string %} | | | {{- content }} | | | {%- elif content is iterable and content is not mapping %} | | | {%- for item in content %} | | | {%- if 'image' in item or 'image url' in item or item.type == 'image' %} | | | {%- if is system content %} | | | {{- raise exception 'System message cannot contain images.' }} | | | {%- endif %} | | | {%- if do vision count %} | | | {%- set image count.value = image count.value + 1 %} | | | {%- endif %} | | | {%- if add vision id is defined and add vision id %} | | | {{- 'Picture ' ~ image count.value ~ ': ' }} | | | {%- endif %} | | | {{- '<|vision start| <|image pad| <|vision end| ' }} | | | {%- elif 'video' in item or item.type == 'video' %} | | | {%- if is system content %} | | | {{- raise exception 'System message cannot contain videos.' }} | | | {%- endif %} | | | {%- if do vision count %} | | | {%- set video count.value = video count.value + 1 %} | | | {%- endif %} | | | {%- if add vision id is defined and add vision id %} | | | {{- 'Video ' ~ video count.value ~ ': ' }} | | | {%- endif %} | | | {{- '<|vision start| <|video pad| <|vision end| ' }} | | | {%- elif 'text' in item %} | | | {{- item.text }} | | | {%- else %} | | | {{- raise exception 'Unexpected item type in content.' }} | | | {%- endif %} | | | {%- endfor %} | | | {%- elif content is none or content is undefined %} | | | {{- '' }} | | | {%- else %} | | | {{- raise exception 'Unexpected content type.' }} | | | {%- endif %} | | | {%- endmacro %} | | | { - ============================================================================ | | | Q1 public fork : preserve thinking default flipped FALSE → TRUE. | | | Why: upstream's preserve thinking gate at the assistant-rendering site | | | is: | | | {%- if preserve thinking is defined and preserve thinking is true | | | or loop.index0 ns.last query index %} | | | With preserve thinking unset, prior-turn <think blocks assistant | | | turns at indices <= last query index are dropped from history. The | | | model loses its own trace of how it chose tool arguments on prior | | | turns and degenerates after 2-3 multi-turn calls of the same tool. | | | The canonical public bug-report on this exact failure mode for | | | Qwen3.6 is earendil-works/pi 3325 : | | | https://github.com/earendil-works/pi/issues/3325 | | | "Qwen3.6 tool calls loop with empty arguments: qwen-chat-template | | | missing preserve thinking ... After 2-3 turns every tool call has | | | arguments: {}." | | | The Qwen3.6 model card explicitly states verbatim : | | | "Qwen3.6 has been additionally trained to preserve and leverage | | | thinking traces from historical messages ... particularly | | | beneficial for agent scenarios." | | | So this is not just a workaround — preserve thinking=true is the | | | model-card-recommended setting for agentic harnesses. The public | | | fork makes it the default. | | | Recover upstream behavior: pass preserve thinking=false explicitly. | | | ============================================================================ - } | | | {%- if preserve thinking is not defined %} | | | {%- set preserve thinking = true %} | | | {%- endif %} | | | { - Q5 / Q6 public fork : both gated by kwargs, default true. See the | | | patch sites below for the full rationale and citations. - } | | | {%- if unwrap tool envelope is not defined %} | | | {%- set unwrap tool envelope = true %} | | | {%- endif %} | | | {%- if verbose tool instructions is not defined %} | | | {%- set verbose tool instructions = true %} | | | {%- endif %} | | | {%- if not messages %} | | | {{- raise exception 'No messages provided.' }} | | | {%- endif %} | | | { - ============================================================================ | | | Q2 public fork : developer role accepted as an alias for system . | | | Upstream's role check in the index-0 system handling AND in the | | | main message loop only accepts system ; a developer role | | | raises "Unexpected message role" and crashes the request. | | | Modern coding harnesses opencode, Claude Code, openclaw, Continue | | | emit a developer role for reasoning-capable models, following | | | OpenAI's Responses API convention. This causes the entire request | | | to fail when pointed at a stock Qwen3.6 server. | | | Reference gist documenting the bite for OpenCode + Qwen3.5 : | | | https://gist.github.com/sudoingX/c2facf7d8f7608c65c1024ef3b22d431 | | | Below: we normalize the index-0 role for the upcoming system-block | | | decision, then in the main message loop we treat both as system. | | | The change is invisible for inputs that only use system . | | | ============================================================================ - } | | | {%- if tools and tools is iterable and tools is not mapping %} | | | {{- '<|im start| system\n' }} | | | {{- " Tools\n\nYou have access to the following functions:\n\n<tools " }} | | | {%- for tool in tools %} | | | {{- "\n" }} | | | { - Q5 public fork : unwrap the OpenAI envelope. | | | Background: harnesses speaking OpenAI tool-call protocol send | | | tool definitions wrapped in {"type":"function","function":{...}}. | | | Upstream passes the WHOLE wrapper through tool | tojson , | | | emitting an extra layer the model has to mentally peel off, | | | and wasting ~12 tokens per tool. | | | Qwen's own most recent coder model unwraps this envelope in | | | its canonical chat template: | | | https://huggingface.co/Qwen/Qwen3-Coder-Next/blob/main/chat template.jinja | | | lines 35-37: {%- if tool.function is defined %}{%- set tool = | | | tool.function %}{%- endif %} . | | | Qwen3.6-27B's upstream template predates that change; this | | | patch backports the unwrap behavior so Qwen3.6 sees the same | | | tool format Qwen3-Coder-Next was trained on. | | | Recover upstream behavior: pass unwrap tool envelope=false. - } | | | {%- if unwrap tool envelope and tool.function is defined %} | | | {{- tool.function | tojson }} | | | {%- else %} | | | {{- tool | tojson }} | | | {%- endif %} | | | {%- endfor %} | | | {{- "\n</tools " }} | | | { - Q6 public fork : strengthened IMPORTANT instructions block. | | | Upstream's IMPORTANT block has 4 bullets. The strengthened | | | version adds three bullets that address documented public Qwen | | | coder failure modes: | | | - "Do NOT omit the opening <tool call tag": | | | https://github.com/QwenLM/Qwen3-Coder/issues/475 | | | - "MUST be at the very beginning of a new line, with NO leading | | | spaces or indentation": | | | https://github.com/block/goose/issues/6883 | | | - "Do NOT nest <tool call blocks inside one another": | | | same 6883 + Roo Code custom-XML interaction patterns | | | These bullets are pure additive guidance to the model; they | | | don't change tool-call wire-format behavior for well-formed | | | outputs, but they reduce error rates on the documented edge | | | cases. | | | Recover upstream behavior: pass verbose tool instructions=false. - } | | | {%- if verbose tool instructions %} | | | {{- '\n\nIf you choose to call a function ONLY reply in the following format with NO suffix:\n\n<tool call \n<function=example function name \n<parameter=example parameter 1 \nvalue 1\n</parameter \n<parameter=example parameter 2 \nThis is the value for the second parameter\nthat can span\nmultiple lines\n</parameter \n</function \n</tool call \n\n<IMPORTANT \nReminder:\n- Function calls MUST follow the specified format: an inner <function=... </function block must be nested within <tool call </tool call XML tags.\n- Do NOT omit the opening <tool call tag. Every function call MUST be wrapped in a complete <tool call ...</tool call block.\n- The <tool call and <function tags MUST be at the very beginning of a new line, with NO leading spaces or indentation.\n- Required parameters MUST be specified.\n- To call multiple functions, output a separate, completely closed <tool call </tool call block for EACH function. Do NOT nest <tool call blocks inside one another.\n- You may provide reasoning inside <think ...</think blocks BEFORE the <tool call , but NOT after. After a tool call there must be NO suffix on the same turn.\n- If no function call is needed, answer the question normally and do not mention function calls.\n</IMPORTANT ' }} | | | {%- else %} | | | {{- '\n\nIf you choose to call a function ONLY reply in the following format with NO suffix:\n\n<tool call \n<function=example function name \n<parameter=example parameter 1 \nvalue 1\n</parameter \n<parameter=example parameter 2 \nThis is the value for the second parameter\nthat can span\nmultiple lines\n</parameter \n</function \n</tool call \n\n<IMPORTANT \nReminder:\n- Function calls MUST follow the specified format: an inner <function=... </function block must be nested within <tool call </tool call XML tags\n- Required parameters MUST be specified\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\n</IMPORTANT ' }} | | | {%- endif %} | | | { - Q2 public fork : accept developer role at index 0. - } | | | {%- if messages 0 .role == 'system' or messages 0 .role == 'developer' %} | | | {%- set content = render content messages 0 .content, false, true |trim %} | | | {%- if content %} | | | {{- '\n\n' + content }} | | | {%- endif %} | | | {%- endif %} | | | {{- '<|im end| \n' }} | | | {%- else %} | | | { - Q2 public fork : accept developer role at index 0. - } | | | {%- if messages 0 .role == 'system' or messages 0 .role == 'developer' %} | | | {%- set content = render content messages 0 .content, false, true |trim %} | | | {{- '<|im start| system\n' + content + '<|im end| \n' }} | | | {%- endif %} | | | {%- endif %} | | | { - last query index walk identical to upstream . When preserve thinking=true | | | the public fork's default , the index produced here is not consulted — | | | the assistant-render guard below only checks preserve thinking first. - } | | | {%- set ns = namespace multi step tool=true, last query index=messages|length - 1 %} | | | {%- for message in messages ::-1 %} | | | {%- set index = messages|length - 1 - loop.index0 %} | | | {%- if ns.multi step tool and message.role == "user" %} | | | {%- set content = render content message.content, false |trim %} | | | {%- if not content.startswith '<tool response ' and content.endswith '</tool response ' %} | | | {%- set ns.multi step tool = false %} | | | {%- set ns.last query index = index %} | | | {%- endif %} | | | {%- endif %} | | | {%- endfor %} | | | {%- if ns.multi step tool %} | | | {{- raise exception 'No user query found in messages.' }} | | | {%- endif %} | | | {%- for message in messages %} | | | {%- set content = render content message.content, true |trim %} | | | {%- if message.role == "system" or message.role == "developer" %} | | | { - Q2 public fork : both roles are valid at the start; upstream | | | rejected developer here. The system block was already rendered | | | above; nothing to emit per-message. - } | | | {%- if not loop.first %} | | | {{- raise exception 'System/developer message must be at the beginning.' }} | | | {%- endif %} | | | {%- elif message.role == "user" %} | | | {{- '<|im start| ' + message.role + '\n' + content + '<|im end| ' + '\n' }} | | | {%- elif message.role == "assistant" %} | | | { - ---------------------------------------------------------------- | | | Q4 public fork : robust </think variant handling + unclosed- | | | think rescue. | | | Upstream's content parsing only recognizes </think , and only | | | when it appears with a properly opened <think somewhere | | | earlier in the content. Three documented failure modes leak: | | | - The model emits </thinking long form — upstream treats | | | the entire content as non-reasoning text, then <think and | | | </thinking literals leak into the model's view of history. | | | - Whitespace variants </ think and </think happen with | | | some quantization runtimes especially older llama.cpp builds . | | | - <tool call appears INSIDE an unclosed <think block the | | | model started reasoning, decided to call a tool, and forgot | | | to close the think block first . | | | The Ollama equivalent of this bug: | | | https://github.com/ollama/ollama/issues/14493 | | | "tool calls in the Qwen 3 and Qwen 3.5 model families would | | | not be parsed correctly if emitted during thinking" | | | fixed in Ollama 0.17.3 . | | | The Qwen3-Coder equivalent model omitting opening tag : | | | https://github.com/QwenLM/Qwen3-Coder/issues/475 | | | Q4 handles all four cases. The strict-improvement contract: | | | for any input upstream parses correctly only </think , | | | properly closed , behavior here is identical. | | | ---------------------------------------------------------------- } | | | {%- set reasoning content = '' %} | | | {%- if message.reasoning content is string %} | | | {%- set reasoning content = message.reasoning content %} | | | {%- else %} | | | {%- set think end = '' %} | | | {%- if '</think ' in content %} | | | {%- set think end = '</think ' %} | | | {%- elif '</thinking ' in content %} | | | {%- set think end = '</thinking ' %} | | | {%- elif '</ think ' in content %} | | | {%- set think end = '</ think ' %} | | | {%- elif '</think ' in content %} | | | {%- set think end = '</think ' %} | | | {%- endif %} | | | {%- if think end %} | | | {%- set parts = content.split think end %} | | | {%- set reasoning content = parts 0 %} | | | {%- set content = parts 1: | join think end %} | | | {%- if '<think ' in reasoning content %} | | | {%- set reasoning content = reasoning content.split '<think ' 1: | join '<think ' %} | | | {%- endif %} | | | {%- elif '<think ' in content %} | | | { - Unclosed think; rescue when followed by <tool call | | | ollama 14493 pattern . - } | | | {%- set prefix = content.split '<think ' 0 %} | | | {%- set think part = content.split '<think ' 1: | join '<think ' %} | | | {%- if '<tool call ' in think part %} | | | {%- set reasoning content = think part.split '<tool call ' 0 %} | | | {%- set content = prefix ~ '\n<tool call ' ~ think part.split '<tool call ' 1: | join '<tool call ' %} | | | {%- else %} | | | {%- set reasoning content = think part %} | | | {%- set content = prefix %} | | | {%- endif %} | | | {%- endif %} | | | {%- endif %} | | | {%- set reasoning content = reasoning content | trim %} | | | {%- set content = content | trim %} | | | { - Strip any leaked <tool call text from content; real tool calls | | | come from the dedicated field. Identical to upstream's intent | | | but expressed inline rather than relying on upstream's regex. - } | | | {%- if message.tool calls and message.tool calls is iterable and message.tool calls is not mapping %} | | | {%- if '<tool call ' in content %} | | | {%- set content = content.split '<tool call ' 0 | trim %} | | | {%- endif %} | | | {%- endif %} | | | { - Reasoning-emission gate. Mirrors upstream's structure exactly, | | | but with the Q1 default flip in effect: preserve thinking | | | defaults true, so prior-turn <think blocks survive. - } | | | {%- if preserve thinking is defined and preserve thinking is true or loop.index0 ns.last query index %} | | | {{- '<|im start| ' + message.role + '\n<think \n' + reasoning content + '\n</think \n\n' + content }} | | | {%- else %} | | | {{- '<|im start| ' + message.role + '\n' + content }} | | | {%- endif %} | | | {%- if message.tool calls and message.tool calls is iterable and message.tool calls is not mapping %} | | | {%- for tool call in message.tool calls %} | | | {%- if tool call.function is defined %} | | | {%- set tool call = tool call.function %} | | | {%- endif %} | | | {%- if loop.first %} | | | {%- if content|trim %} | | | {{- '\n\n<tool call \n<function=' + tool call.name + ' \n' }} | | | {%- else %} | | | {{- '<tool call \n<function=' + tool call.name + ' \n' }} | | | {%- endif %} | | | {%- else %} | | | {{- '\n<tool call \n<function=' + tool call.name + ' \n' }} | | | {%- endif %} | | | {%- if tool call.arguments is defined %} | | | { - ---------------------------------------------------- | | | Q3 public fork : debuggable raise on string args. | | | Upstream uses tool call.arguments | items line 120 | | | of upstream/chat template.jinja . When arguments | | | is a JSON-encoded STRING — which is the wire-format | | | the OpenAI spec defines, and what some harness | | | adapters notably the Vercel AI SDK used by | | | opencode hand back to the harness — .items | | | raises: | | | "Can only get item pairs from a mapping" | | | which is impossible to debug without reading the | | | Jinja runtime source. | | | Q3 type-checks first and raises a clear error that | | | names the bug surface and links to the canonical | | | discussion. Harnesses MUST deserialize the JSON- | | | encoded arguments string exactly once on ingest | | | and store the resulting dict. See: | | | https://github.com/earendil-works/pi/issues/3325 | | | https://github.com/anomalyco/opencode/issues/24264 | | | For inputs where arguments is already a dict the | | | well-formed case , behavior is identical to upstream. | | | ---------------------------------------------------- } | | | {%- if tool call.arguments is mapping %} | | | {%- for args name, args value in tool call.arguments|items %} | | | {{- '<parameter=' + args name + ' \n' }} | | | {%- set args value = args value | string if args value is string else args value | tojson | safe %} | | | {{- args value }} | | | {{- '\n</parameter \n' }} | | | {%- endfor %} | | | {%- elif tool call.arguments is string %} | | | {{- raise exception | | | "custom pub chat template qwen36: " | | | "tool call.arguments must be a JSON object " | | | " mapping . Got a string. This is almost " | | | "always the harness handing back a JSON-" | | | "encoded STRING rather than the deserialized " | | | "object common with Vercel AI SDK . " | | | "Deserialize once on ingest and store the " | | | "object. See: " | | | "github.com/earendil-works/pi/issues/3325" | | | }} | | | {%- endif %} | | | {%- endif %} | | | {{- '</function \n</tool call ' }} | | | {%- endfor %} | | | {%- endif %} | | | {{- '<|im end| \n' }} | | | {%- elif message.role == "tool" %} | | | {%- if loop.previtem and loop.previtem.role = "tool" %} | | | {{- '<|im start| user' }} | | | {%- endif %} | | | {{- '\n<tool response \n' }} | | | {{- content }} | | | {{- '\n</tool response ' }} | | | {%- if not loop.last and loop.nextitem.role = "tool" %} | | | {{- '<|im end| \n' }} | | | {%- elif loop.last %} | | | {{- '<|im end| \n' }} | | | {%- endif %} | | | {%- else %} | | | {{- raise exception 'Unexpected message role.' }} | | | {%- endif %} | | | {%- endfor %} | | | {%- if add generation prompt %} | | | {{- '<|im start| assistant\n' }} | | | {%- if enable thinking is defined and enable thinking is false %} | | | {{- '<think \n\n</think \n\n' }} | | | {%- else %} | | | {{- '<think \n' }} | | | {%- endif %} | | | {%- endif %} |