@miller-tech/uap 1.216.0 → 1.218.0

package/docs/guides/QWEN36_LLAMACPP.md

@@ -46,12 +46,18 @@ llama-server \
   --ctx-size    <PER TIER> \
   --cache-type-k <q4_0|q8_0> --cache-type-v <q4_0|q8_0> \   # quantize KV cache to save VRAM
   --flash-attn on \             # faster + less VRAM
-  --jinja \                     # use the model's chat template (REQUIRED for tool calls)
+  --jinja \                     # enable chat-template-driven tool calls (REQUIRED)
+  --chat-template-file tools/agents/config/qwen-sharp.jinja \
   --parallel 1
 ```
 
-> **`--jinja` / the chat template is non-negotiable for agent use.** Qwen3.6
-> emits *native* OpenAI tool calls, but only when its chat template is active.
+> **`--jinja` plus an explicit `--chat-template-file` is non-negotiable for agent
+> use.** Qwen emits tool calls only when a chat template is active, and `--jinja`
+> alone uses the model's *embedded* template — which still hard-raises
+> `'System message must be at the beginning.'`, breaking any request where UAP's
+> memory or context-injection layers add a system message mid-conversation. Always
+> pass `--chat-template-file` (see
+> [Chat template: `qwen-sharp.jinja`](#chat-template-qwen-sharpjinja-all-qwen-models)).
 > If tools silently never fire, that's the cause — run `uap tool-calls setup`
 > to install/repair the template, and `uap tool-calls status` to check.
 
@@ -74,6 +80,195 @@ Speed extras (any tier): `--flash-attn on` (always), and if your build supports
 it, **self-speculation / MTP** (the `*-MTP.gguf` model, or `--draft-*` flags with
 a tiny draft model) for materially faster decode.
 
+## Measured: Qwen3.8-27B (dense) + native `draft-mtp`, RTX 3090 24 GB
+
+Production switched 2026-08-20 to a **dense** qwen35-arch model
+(`Qwen3.8-27B-UD-IQ4_XS.gguf`, ~13.3 GiB on disk) with a **native MTP head
+baked into the checkpoint** (`blk.64.nextn.*` tensors) — no separate
+`*-MTP.gguf` or `-md` draft file needed; `--spec-type draft-mtp` reads the
+head straight from `--model`. Built at llama.cpp master (`f466cfa38`,
+2026-08-20). See `config/llama-profiles/qwen38-27b-mtp.env` for the full
+resolved profile.
+
+Measured VRAM on this card (`q4_0`/`q4_0` KV, `-ngl 99`, `--parallel 2`,
+`--flash-attn on`, no vision projector):
+
+| Per-slot ctx (`--ctx-size` total) | VRAM used | VRAM free | Notes |
+|---|---|---|---|
+| 65536 (`131072` total) | ~19.8 GiB | ~4.3 GiB | Comfortable headroom. |
+| 131072 (`262144` total) | ~23.15 GiB | ~0.95 GiB | Verified stable under 2 concurrent long-generation requests (60-91% draft acceptance, no OOM), but headroom is thin on a 24 GiB card with a history of kernel OOM-kills reaping this service under unrelated host memory pressure. If instability appears under real concurrent load, drop back to 65536/slot first. |
+
+Doubling `--ctx-size` roughly doubles KV cache size, not the model footprint
+(weights stay fixed at ~13.3 GiB) — so the free-VRAM gap shrinks fast once
+weights + KV together approach the card's total. Always re-check
+`nvidia-smi`/`/props` after raising ctx on this card rather than assuming the
+previous tier's headroom scales.
+
+## Chat template: `qwen-sharp.jinja` (all Qwen models)
+
+Every Qwen launch path here runs
+[`tools/agents/config/qwen-sharp.jinja`](../../tools/agents/config/qwen-sharp.jinja),
+vendored verbatim from
+[peculiar-ragdoll/Qwen-Sharp-Chat-Templates](https://huggingface.co/peculiar-ragdoll/Qwen-Sharp-Chat-Templates)
+(`template_version` `qwen3.8-froggeric-v22.1`, upstream revision
+`3dc34df52c63dd22ada21f96435e069deaa8d7da`). It targets Qwen 3.5 / 3.6 / 3.8 and
+replaced `qwen3.5-enhanced.jinja` on 2026-08-20. The file's sha256 is asserted in
+`test/chat-template-system-message.test.ts`, so a silent re-vendor from a changed
+upstream fails CI rather than quietly changing the prompt contract.
+
+### What it fixes
+
+`qwen3.5-enhanced.jinja` hard-raised `'No user query found in messages.'` — a
+server-side failure for the entire request — whenever its backwards scan found no
+user turn that wasn't a `<tool_response>`. Two shapes that occur in ordinary agent
+loops trigger it: a continuation whose only user turns are tool results, and a
+`system`+`assistant` transcript with no user turn at all. `qwen-sharp` renders both.
+
+This is *not* the mid-conversation-system-message case — both templates always
+handled that. Earlier notes in this repo attributed the raise to the wrong cause.
+
+### What stays compatible
+
+Verified by rendering both templates over a fixture table and diffing the output,
+not by reading the source:
+
+* the tool-call contract the model is instructed to **emit** is unchanged —
+  `<tool_call><function=NAME><parameter=K>V</parameter></function></tool_call>`,
+  which `anthropic_proxy.py`'s `_HERMES_FUNCTION_RE` / `_TOOL_CALL_XML_RE` parse
+  unchanged;
+* the generation prompt matches the old template in **both** `enable_thinking`
+  states, so `--reasoning auto --reasoning-format deepseek` is unaffected;
+* a mid-conversation system message renders inline in both.
+
+The template ships **both** a JSON and an XML tool-call emitter and picks one at
+render time via `_tool_format`. That is why the tests assert on rendered output: a
+grep for the XML emitter's source passes even if the default flips to JSON and
+every tool call on the box silently changes wire format.
+
+### What actually differs — known and accepted
+
+* **History whitespace.** Replaying a *prior* tool call with string-form
+  `arguments` (the proxy's path — it `json.dumps` them) emitted `}\n</function>`
+  before and `}</function>` now; the separator between two tool calls in one
+  assistant turn went `\n` → `\n\n`. Neither changes what the parsers accept.
+* **The injected tool-calling instruction block was rewritten upstream.** It now
+  requires a `<think>` block before each tool call and forbids prose before the
+  call. Behavioural and deliberate, but it is more than a formatting change.
+* **A system message containing an image now raises.** Unreachable through the
+  proxy, which flattens `system` to a text string before dispatch; direct-to-`:8080`
+  OpenAI clients can hit it.
+* **Unknown message roles no longer raise** — they render as `[role]: content`.
+  A malformed request now proceeds quietly where it used to fail loudly.
+* **Prior `<think>` blocks are retained across turns.** Upstream sells this as a
+  prompt-cache win. On the proxy path it is a **no-op**: the proxy already strips
+  `<think>` from prior assistant turns on purpose ("Stripping breaks the copy
+  cycle"). On the direct path it re-introduces exactly that cycle — pass
+  `chat_template_kwargs.preserve_thinking=false` there if it shows up.
+
+`chat_template_kwargs.reasoning_effort` accepts
+`none|minimal|low|medium|high|xhigh|max`; note `none` forces thinking off,
+overriding `enable_thinking=true`.
+
+Do **not** switch a Qwen profile to `embedded` — the model's own baked-in template
+still carries the `'System message must be at the beginning.'` raise.
+
+## Speculative decoding: DFlash2 is the default
+
+`config/llama-profiles/qwen38-27b-dflash2.env` is the default profile as of
+2026-08-20. `qwen38-27b-mtp.env` is kept as the fallback — same model, same
+context, same slot count; only the speculative-decoding config differs.
+
+Measured on this card (RTX 3090, sm_86, temp 0, 4 prompts, 1032 tokens per arm,
+**same binary across arms** so the spec config is the only variable):
+
+| arm | aggregate tok/s | accept rate | tokens/verify step | |
+|---|---|---|---|---|
+| no speculation, `-np 1` | 41.3 | — | 1.00 | baseline |
+| `draft-mtp` n-max 2, `-np 1` | 59.3 | 93.6% | 2.68 | |
+| **DFlash2 n-max 4, `-np 1`** | **68.7** | 67.4% | 3.68 | **1.16×** |
+| `draft-mtp` n-max 2, `-np 2` | 66.0 | 92.5% | 2.67 | |
+| **DFlash2 n-max 4, `-np 2`** | **89.8** | 66.5% | 3.66 | **1.36×** |
+
+The `-np 2` rows are two *genuinely concurrent* requests — the production shape.
+Re-verified end-to-end through `run-llama-server-continuity.sh` with the shipped
+profile: **92.6 tok/s**, tool calls parsing, peak 23459 MiB.
+
+**Read `1.36×` as a blend, not a constant.** That run pushed 4 prompts through a
+2-slot pool, so the last prompt always finishes alone and the wall covers two
+different regimes. Decomposing it from the server's own `launch_slot_`/`release`
+timeline:
+
+| regime | `draft-mtp` | DFlash2 n-max 4 | ratio |
+|---|---|---|---|
+| both slots decoding | 76.4 tok/s | 130.3 tok/s | **1.71×** |
+| one request in flight on the 2-slot server | 57.9 tok/s | 60.4 tok/s | **1.04×** |
+| the measured 4-prompt burst (what `1.36×` reports) | 66.0 | 89.8 | 1.36× |
+
+So the gain is real but duty-cycle dependent: it is ~1.7× while both slots are
+busy and close to nothing when only one request is in flight. Add prompts to the
+burst and the headline drifts up toward 1.71×; remove them and it drifts down.
+The `-np 1` row above is a `--parallel 1` server, which this service never runs —
+for the single-request case on the real 2-slot server use the 1.04–1.2× range,
+not 1.16×.
+
+Two caveats on the numbers, both unclosed:
+
+* **Every measurement was taken at `temperature 0.0`**, while this profile serves
+  at `LLAMA_TEMP=0.7 / TOP_P=0.8 / TOP_K=20`. Greedy is the most favourable
+  setting for speculative decoding, and it flatters a 4-token block more than a
+  2-token one, so `1.36×` is an upper bound for the sampling that actually runs.
+* **Acceptance is strongly prompt-dependent for DFlash2 and flat for MTP.**
+  Across the six prompts ever run, DFlash2 ranged 0.40–0.95 and `draft-mtp` sat
+  in a 0.91–0.97 band. The two lowest-acceptance prompts (prose explanation) were
+  never run against `draft-mtp`, so the regime where DFlash2 could lose is
+  untested. `n-max 7 < n-max 4` is direct evidence that drafting cost dominates
+  once acceptance falls.
+
+Two results here are worth keeping, because both contradict what the upstream
+material predicts:
+
+* **DFlash2 does not degrade under concurrency on this card — it improves.** The
+  PR thread reports throughput collapsing at `-np >= 2`; here the advantage grows
+  from 1.16× to 1.36× going from one slot to two. Do not carry that assumption to
+  other hardware in either direction; measure it.
+* **`n-max 4` beats `n-max 7`** (76.7 vs 69.8 tok/s), even though n-max 7 accepts
+  a longer block (4.86 vs 3.87 tokens/step). Acceptance falls from 72% to 55% as
+  the block lengthens, so the extra drafting outruns what it buys. The drafter
+  card's published acceptance length of 5.39 is real but is not the throughput
+  optimum.
+
+### The cost, and when to roll back
+
+Peak VRAM in the production shape is **23459 / 24576 MiB — about 1.1 GiB free**.
+There is **no paired measurement of `draft-mtp` taken the same way**, so treat the
+drafter's marginal cost as un-measured rather than as the ~0.8 GiB the weights
+imply: the table above reports ~0.95 GiB free for `draft-mtp` at this same tier,
+which cannot both be true and leave DFlash2 roomier. Re-measure both profiles
+with one harness before relying on either figure. Note also that 23459 was
+sampled with `nvidia-smi` on a 1 s loop over a ~10 s run whose longest prefill
+was 540 tokens — it never exercised a long-context prefill on two slots, which is
+the real peak event at 131072/slot. The drafter's weights (Q4_K_M,
+1.1 GB) are essentially the whole difference: its KV cache stays around 80 MiB
+because it carries its own 2048-token sliding window rather than scaling with the
+target's 131072/slot. That is what lets this fit without cutting context or slots
+— and it is why `LLAMA_DRAFT_CTX_SIZE` must stay unset.
+
+This card has a history of *global* kernel OOM events selecting this service for
+its RSS. The GPU side is now thinner too. **If the server starts failing to load
+or dying, switch `~/.config/uap/llama-server.env` back to `qwen38-27b-mtp.env`
+before changing anything else** — that is what the fallback profile is for.
+
+### The engine is an unmerged PR
+
+[llama.cpp#27342](https://github.com/ggml-org/llama.cpp/pull/27342) is **open**.
+The build is a cherry-pick of `5ecbe1ac1` onto `f466cfa38` (zero conflicts),
+vendored as `config/llama-patches/dflash2-pr27342.patch` so it is reproducible if
+the build worktree is lost. The merged `draft-dflash` **v1** engine cannot
+substitute: a v2 checkpoint declares the same `general.architecture = "dflash"`,
+so auto-detect selects `draft-dflash` and then fails on the 23 conv/selector
+tensors and `dflash.conv_*` / `dflash.selector_*` metadata keys v1 has no loader
+for. Return to plain upstream master once the PR lands.
+
 ## Point UAP at it
 
 ```bash

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@miller-tech/uap",
-  "version": "1.216.0",
+  "version": "1.218.0",
   "description": "Autonomous AI agent memory system with CLAUDE.md protocol enforcement",
   "type": "module",
   "main": "dist/index.js",

package/tools/agents/README.md

@@ -13,9 +13,19 @@ This directory contains tools and configurations for fixing Qwen3.5 tool calling
 
 ## Files
 
+### `config/qwen-sharp.jinja`
+
+The template every Qwen profile runs (Qwen 3.5 / 3.6 / 3.8), vendored verbatim from
+[peculiar-ragdoll/Qwen-Sharp-Chat-Templates](https://huggingface.co/peculiar-ragdoll/Qwen-Sharp-Chat-Templates)
+and sha256-pinned in `test/chat-template-system-message.test.ts`. See
+[docs/guides/QWEN36_LLAMACPP.md](../../docs/guides/QWEN36_LLAMACPP.md#chat-template-qwen-sharpjinja-all-qwen-models)
+for what it changes and what stays wire-compatible.
+
+`uap tool-calls setup` installs it automatically under any `qwen*` profile.
+
 ### `config/chat_template.jinja`
 
-The core fix: a patched Jinja2 template for Qwen3.5 that adds conditional wrappers around tool call argument iteration.
+The generic (non-Qwen) fallback: a patched Jinja2 template for Qwen3.5 that adds conditional wrappers around tool call argument iteration.
 
 **Key Fix (line 138-144):**
 
@@ -142,7 +152,7 @@ cp tools/agents/scripts/*.py tools/agents/scripts/
 ```bash
 ./llama-server \
   --model ~/models/Qwen3.5-35B-Instruct-Q4_K_M.gguf \
-  --chat-template-file tools/agents/config/chat_template.jinja \
+  --chat-template-file tools/agents/config/qwen-sharp.jinja \
   --jinja \
   --port 8080 \
   --ctx-size 262144 \
@@ -161,7 +171,7 @@ cp tools/agents/scripts/*.py tools/agents/scripts/
 
 ```bash
 mkdir -p ~/.opencode/agent
-cp tools/agents/config/chat_template.jinja ~/.opencode/agent/
+cp tools/agents/config/qwen-sharp.jinja ~/.opencode/agent/
 ```
 
 **2. Update `.opencode/config.json`:**

package/tools/agents/config/qwen-sharp.jinja

@@ -0,0 +1,374 @@
+{%- set template_version = "qwen3.8-froggeric-v22.1" %}
+{%- set _tool_format = tool_call_format if tool_call_format is defined else 'xml' %}
+{%- set image_count = namespace(value=0) %}
+{%- set video_count = namespace(value=0) %}
+{%- set add_vision_id = add_vision_id if add_vision_id is defined else false %}
+{%- set enable_thinking = enable_thinking if enable_thinking is defined else true %}
+{%- set auto_disable_thinking_with_tools = auto_disable_thinking_with_tools if auto_disable_thinking_with_tools is defined else false %}
+{%- if preserve_reasoning is defined and preserve_reasoning is not none %}
+    {%- set _preserve_thinking = preserve_reasoning %}
+{%- elif preserve_thinking is defined and preserve_thinking is not none %}
+    {%- set _preserve_thinking = preserve_thinking %}
+{%- else %}
+    {%- set _preserve_thinking = true %}
+{%- endif %}
+{%- set max_tool_arg_chars = max_tool_arg_chars if max_tool_arg_chars is defined else 0 %}
+{%- set max_tool_response_chars = max_tool_response_chars if max_tool_response_chars is defined else 0 %}
+{%- set _has_tools = (tools is defined and tools and tools is iterable and tools is not mapping) %}
+{%- set _effort_raw = (reasoning_effort | string | lower) if reasoning_effort is defined and reasoning_effort is not none else 'medium' %}
+{%- set _initial_thinking = enable_thinking %}
+{%- set _initial_effort = 'medium' %}
+{%- if _effort_raw == 'none' %}
+    {%- set _initial_thinking = false %}
+    {%- set _initial_effort = 'medium' %}
+{%- elif _effort_raw == 'minimal' or _effort_raw == 'low' %}
+    {%- set _initial_effort = 'low' %}
+{%- elif _effort_raw == 'high' or _effort_raw == 'xhigh' or _effort_raw == 'max' %}
+    {%- set _initial_effort = 'xhigh' %}
+{%- else %}
+    {%- set _initial_effort = 'medium' %}
+{%- endif %}
+{%- set ns_state = namespace(thinking=_initial_thinking, effort=_initial_effort) %}
+{%- if auto_disable_thinking_with_tools and _has_tools %}
+    {%- set ns_state.thinking = false %}
+{%- endif %}
+{%- for msg in messages %}
+    {%- if msg.role == 'system' or msg.role == 'developer' or msg.role == 'user' %}
+        {%- if msg.content is string %}
+            {%- if '<|think_off|>' in msg.content %}
+                {%- set ns_state.thinking = false %}
+            {%- elif '<|think_on|>' in msg.content %}
+                {%- set ns_state.thinking = true %}
+            {%- elif '<|think_xhigh|>' in msg.content or '<|think_high|>' in msg.content %}
+                {%- set ns_state.thinking = true %}
+                {%- set ns_state.effort = 'xhigh' %}
+            {%- elif '<|think_low|>' in msg.content or '<|think_minimal|>' in msg.content %}
+                {%- set ns_state.thinking = true %}
+                {%- set ns_state.effort = 'low' %}
+            {%- elif '<|think_medium|>' in msg.content %}
+                {%- set ns_state.thinking = true %}
+                {%- set ns_state.effort = 'medium' %}
+            {%- endif %}
+        {%- elif msg.content is iterable and msg.content is not mapping %}
+            {%- for item in msg.content %}
+                {%- if item is string %}
+                    {%- set _item_text = item %}
+                {%- elif item is mapping and 'text' in item and item.text is string %}
+                    {%- set _item_text = item.text %}
+                {%- else %}
+                    {%- set _item_text = '' %}
+                {%- endif %}
+                {%- if _item_text %}
+                    {%- if '<|think_off|>' in _item_text %}
+                        {%- set ns_state.thinking = false %}
+                    {%- elif '<|think_on|>' in _item_text %}
+                        {%- set ns_state.thinking = true %}
+                    {%- elif '<|think_xhigh|>' in _item_text or '<|think_high|>' in _item_text %}
+                        {%- set ns_state.thinking = true %}
+                        {%- set ns_state.effort = 'xhigh' %}
+                    {%- elif '<|think_low|>' in _item_text or '<|think_minimal|>' in _item_text %}
+                        {%- set ns_state.thinking = true %}
+                        {%- set ns_state.effort = 'low' %}
+                    {%- elif '<|think_medium|>' in _item_text %}
+                        {%- set ns_state.thinking = true %}
+                        {%- set ns_state.effort = 'medium' %}
+                    {%- endif %}
+                {%- endif %}
+            {%- endfor %}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- set reasoning_instructions = '' %}
+{%- if ns_state.thinking %}
+    {%- if ns_state.effort == 'xhigh' %}
+        {%- set reasoning_instructions = 'Reasoning effort is set to xhigh. Please think carefully through the task, validate key assumptions, consider plausible alternatives, and prioritize correctness, consistency, and clarity in the final answer.' %}
+    {%- elif ns_state.effort == 'low' %}
+        {%- set reasoning_instructions = 'Reasoning effort is set to low. Keep your thinking brief and focused, moving directly to the conclusion without unnecessary elaboration.' %}
+    {%- endif %}
+{%- endif %}
+{%- 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 item is mapping %}
+                {%- if item.type == 'image' or 'image' in item or 'image_url' in item %}
+                    {%- 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 %}
+                        {{- 'Picture ' ~ image_count.value ~ ': ' }}
+                    {%- endif %}
+                    {{- '<|vision_start|><|image_pad|><|vision_end|>' }}
+                {%- elif item.type == 'video' or 'video' in item %}
+                    {%- 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 %}
+                        {{- '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 %}
+            {%- else %}
+                {{- item | string }}
+            {%- endif %}
+        {%- endfor %}
+    {%- elif content is none or content is undefined %}
+        {{- '' }}
+    {%- else %}
+        {{- raise_exception('Unexpected content type.') }}
+    {%- endif %}
+{%- endmacro %}
+{%- if not messages %}
+    {{- raise_exception('No messages provided.') }}
+{%- endif %}
+{%- set _first_role = messages[0].role %}
+{%- if _first_role == 'system' or _first_role == 'developer' %}
+    {%- set _sys_msg = messages[0] %}
+    {%- set _msgs = messages[1:] %}
+{%- else %}
+    {%- set _sys_msg = none %}
+    {%- set _msgs = messages %}
+{%- endif %}
+{%- set _sc = '' %}
+{%- if _sys_msg is not none %}
+    {%- set _sc = render_content(_sys_msg.content, false, true) | trim %}
+    {%- if '<|think_off|>' in _sc %}{%- set _sc = _sc.split('<|think_off|>') | join('') | trim %}{%- endif %}
+    {%- if '<|think_on|>' in _sc %}{%- set _sc = _sc.split('<|think_on|>') | join('') | trim %}{%- endif %}
+    {%- if '<|think_xhigh|>' in _sc %}{%- set _sc = _sc.split('<|think_xhigh|>') | join('') | trim %}{%- endif %}
+    {%- if '<|think_high|>' in _sc %}{%- set _sc = _sc.split('<|think_high|>') | join('') | trim %}{%- endif %}
+    {%- if '<|think_medium|>' in _sc %}{%- set _sc = _sc.split('<|think_medium|>') | join('') | trim %}{%- endif %}
+    {%- if '<|think_low|>' in _sc %}{%- set _sc = _sc.split('<|think_low|>') | join('') | trim %}{%- endif %}
+    {%- if '<|think_minimal|>' in _sc %}{%- set _sc = _sc.split('<|think_minimal|>') | join('') | trim %}{%- endif %}
+{%- endif %}
+{%- set _terse %}
+Answer directly, after thinking. Lead with the answer, then only what it needs to be correct and usable.
+Never: open with preamble or pleasantries; restate the question; add filler transitions; hedge with niceties; or repeat a point you've already made.
+Always: keep essential steps, caveats, uncertainties, and specifics — never drop correctness or a needed warning for brevity. Keep the final answer lean. Use the least structure that conveys it (plain prose when short; lists or code only when they earn their place). If genuinely uncertain, say so and explain why — never omit uncertainty for the sake of brevity.
+If a user request is genuinely ambiguous, ask a sharp question, don't guess.
+{%- endset %}
+{%- if not _sc %}
+    {%- set _sc = _terse | trim %}
+{%- else %}
+    {%- set _sc = (_sc | trim) ~ '\n\n' ~ (_terse | trim) %}
+{%- endif %}
+{%- if _has_tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if reasoning_instructions %}
+        {{- reasoning_instructions + '\n\n' }}
+    {%- endif %}
+    {{- '# Tools\n\nYou have access to the following functions:\n\n<tools>' }}
+    {%- for tool in tools %}
+        {{- '\n' }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- '\n</tools>' }}
+    {%- if _tool_format == 'json' %}
+        {{- '\n\nIf you choose to call a function ONLY reply in the following format with NO suffix:\n\n<think>\nBrief explanation of tool call\n</think>\n<tool_call>\n{"name": "example_function_name", "arguments": {"example_parameter_1": "value_1", "example_parameter_2": "This is the value for the second parameter"}}\n</tool_call>\n\n<IMPORTANT>\nReminder:\n- You can use the <think></think> block to plan your next tool call OR to synthesize data and formulate your final response to the user.\n- ALL explanation and reasoning MUST be placed strictly inside the <think></think> block.\n- Function calls MUST follow the specified format: a single JSON object with "name" and "arguments" keys inside <tool_call></tool_call> XML tags.\n- If you choose to call a tool, you MUST output the <tool_call> block IMMEDIATELY after thinking, with NO conversational text before it.\n- The <tool_call> tag MUST be at the very beginning of a new line, with NO spaces or indentation before it.\n- To call multiple functions, output a separate, completely closed <tool_call></tool_call> block for EACH function. Do NOT nest <tool_call> blocks.\n- If you have all necessary data, provide your final answer directly to the user without any tool call.\n</IMPORTANT>' }}
+    {%- else %}
+        {{- '\n\nIf you choose to call a function ONLY reply in the following format with NO suffix:\n\n<think>\nBrief explanation of tool call\n</think>\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- You can use the <think></think> block to plan your next tool call OR to synthesize data and formulate your final response to the user.\n- ALL explanation and reasoning MUST be placed strictly inside the <think></think> block.\n- Function calls MUST follow the specified format: an inner <function=...></function> block must be nested within <tool_call></tool_call> XML tags.\n- If you choose to call a tool, you MUST output the <tool_call> block IMMEDIATELY after thinking, with NO conversational text before it.\n- The <tool_call> and <function> tags MUST be at the very beginning of a new line, with NO spaces or indentation before them.\n- To call multiple functions, output a separate, completely closed <tool_call></tool_call> block for EACH function. Do NOT nest <tool_call> blocks.\n- If you have all necessary data, provide your final answer directly to the user without any tool call.\n</IMPORTANT>' }}
+    {%- endif %}
+    {%- if _sc %}
+        {{- '\n\n' + _sc }}
+    {%- endif %}
+    {{- '<|im_end|>\n' }}
+{%- else %}
+    {%- if _sc %}
+        {{- '<|im_start|>system\n' + (reasoning_instructions + '\n\n' if reasoning_instructions else '') + _sc + '<|im_end|>\n' }}
+    {%- elif reasoning_instructions %}
+        {{- '<|im_start|>system\n' + reasoning_instructions + '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- set _last_idx = _msgs | length - 1 %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=_last_idx) %}
+{%- for message in _msgs[::-1] %}
+    {%- set index = (_msgs | length - 1) - loop.index0 %}
+    {%- if ns.multi_step_tool and message.role == 'user' %}
+        {%- set _rc = render_content(message.content, false) | trim %}
+        {%- if not (_rc.startswith('<tool_response>') and _rc.endswith('</tool_response>')) %}
+            {%- set ns.multi_step_tool = false %}
+            {%- set ns.last_query_index = index %}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if ns.multi_step_tool %}
+    {%- if _last_idx > 50 %}
+        {%- set ns.last_query_index = _last_idx %}
+    {%- else %}
+        {%- set ns.last_query_index = 0 %}
+    {%- endif %}
+{%- endif %}
+{%- set ns2 = namespace(prev_role='', consecutive_failures=0) %}
+{%- for message in _msgs %}
+    {%- set is_system = (message.role == "system" or message.role == "developer") %}
+    {%- set content = render_content(message.content, true, is_system) | trim %}
+    {%- if is_system or message.role == 'user' %}
+        {%- if '<|think_off|>' in content %}{%- set content = content.split('<|think_off|>') | join('') | trim %}{%- endif %}
+        {%- if '<|think_on|>' in content %}{%- set content = content.split('<|think_on|>') | join('') | trim %}{%- endif %}
+        {%- if '<|think_xhigh|>' in content %}{%- set content = content.split('<|think_xhigh|>') | join('') | trim %}{%- endif %}
+        {%- if '<|think_high|>' in content %}{%- set content = content.split('<|think_high|>') | join('') | trim %}{%- endif %}
+        {%- if '<|think_medium|>' in content %}{%- set content = content.split('<|think_medium|>') | join('') | trim %}{%- endif %}
+        {%- if '<|think_low|>' in content %}{%- set content = content.split('<|think_low|>') | join('') | trim %}{%- endif %}
+        {%- if '<|think_minimal|>' in content %}{%- set content = content.split('<|think_minimal|>') | join('') | trim %}{%- endif %}
+    {%- endif %}
+    {%- if is_system %}
+        {{- '<|im_start|>system\n' + content + '<|im_end|>\n' }}
+    {%- elif message.role == 'user' %}
+        {%- set ns2.consecutive_failures = 0 %}
+        {{- '<|im_start|>user\n' + content + '<|im_end|>\n' }}
+    {%- elif message.role == 'assistant' %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is defined and message.reasoning_content is not none %}
+            {%- if message.reasoning_content is string %}
+                {%- set reasoning_content = message.reasoning_content %}
+            {%- else %}
+                {%- set reasoning_content = message.reasoning_content | string %}
+            {%- endif %}
+        {%- elif message.thinking is defined and message.thinking is not none %}
+            {%- if message.thinking is string %}
+                {%- set reasoning_content = message.thinking %}
+            {%- else %}
+                {%- set reasoning_content = message.thinking | string %}
+            {%- endif %}
+        {%- else %}
+            {%- set _think_end = '' %}
+            {%- if content.startswith('</think>') %}
+                {%- set _think_end = '</think>' %}
+            {%- elif content.startswith('</thinking>') %}
+                {%- set _think_end = '</thinking>' %}
+            {%- elif '\n</think>' in content %}
+                {%- set _think_end = '\n</think>' %}
+            {%- elif '\n</thinking>' in content %}
+                {%- set _think_end = '\n</thinking>' %}
+            {%- elif '\n</ think>' in content %}
+                {%- set _think_end = '\n</ think>' %}
+            {%- elif '\n</think >' in content %}
+                {%- set _think_end = '\n</think >' %}
+            {%- endif %}
+            {%- if _think_end %}
+                {%- if 'thinking' in _think_end %}
+                    {%- set _think_start = '<thinking>' %}
+                {%- else %}
+                    {%- set _think_start = '<think>' %}
+                {%- endif %}
+                {%- set reasoning_content = content.split(_think_end)[0].rstrip('\n') %}
+                {%- if _think_start in reasoning_content %}
+                    {%- set reasoning_content = reasoning_content.split(_think_start)[-1].lstrip('\n') %}
+                {%- endif %}
+                {%- set content = content.split(_think_end)[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- set reasoning_content = reasoning_content | trim %}
+        {%- if (_preserve_thinking or loop.index0 > ns.last_query_index) and reasoning_content %}
+            {{- '<|im_start|>assistant\n<think>\n' + reasoning_content + '\n</think>\n\n' + content }}
+        {%- else %}
+            {{- '<|im_start|>assistant\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls is defined and 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 and tool_call.function is not none %}
+                    {%- set tc = tool_call.function %}
+                {%- else %}
+                    {%- set tc = tool_call %}
+                {%- endif %}
+                {%- set tc_name = tc.name if (tc.name is defined and tc.name is not none) else '' %}
+                {%- if _tool_format == 'json' %}
+                    {%- if not loop.first or content | trim %}
+                        {{- '\n\n' }}
+                    {%- endif %}
+                    {%- set _args = '{}' %}
+                    {%- if tc.arguments is defined and tc.arguments is not none %}
+                        {%- if tc.arguments is mapping %}
+                            {%- set _args = tc.arguments | tojson %}
+                        {%- elif tc.arguments is string and tc.arguments %}
+                            {%- set _args = tc.arguments %}
+                        {%- endif %}
+                    {%- endif %}
+                    {{- '<tool_call>\n{"name": ' }}{{- tc_name | tojson }}{{- ', "arguments": ' }}{{- _args }}{{- '}\n</tool_call>' }}
+                {%- else %}
+                    {%- if loop.first %}
+                        {%- if content | trim %}
+                            {{- '\n\n<tool_call>\n<function=' + tc_name + '>\n' }}
+                        {%- else %}
+                            {{- '<tool_call>\n<function=' + tc_name + '>\n' }}
+                        {%- endif %}
+                    {%- else %}
+                        {{- '\n\n<tool_call>\n<function=' + tc_name + '>\n' }}
+                    {%- endif %}
+                    {%- if tc.arguments is defined and tc.arguments is not none %}
+                        {%- if tc.arguments is mapping %}
+                            {%- for args_name, args_value in tc.arguments.items() %}
+                                {{- '<parameter=' + args_name + '>\n' }}
+                                {%- if args_value is mapping or (args_value is sequence and args_value is not string) %}
+                                    {%- set _av = args_value | tojson %}
+                                {%- else %}
+                                    {%- set _av = args_value | string %}
+                                {%- endif %}
+                                {%- if max_tool_arg_chars > 0 and _av | length > max_tool_arg_chars %}
+                                    {{- _av[:max_tool_arg_chars] + '\n[TRUNCATED - original length ' ~ (_av | length | string) ~ ' chars]' }}
+                                {%- else %}
+                                    {{- _av }}
+                                {%- endif %}
+                                {{- '\n</parameter>\n' }}
+                            {%- endfor %}
+                        {%- elif tc.arguments is string and tc.arguments %}
+                            {{- tc.arguments }}
+                        {%- endif %}
+                    {%- endif %}
+                    {{- '</function>\n</tool_call>' }}
+                {%- endif %}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == 'tool' %}
+        {%- set _content_lower = content | lower %}
+        {%- set _content_head = _content_lower[:80] %}
+        {%- if content | length < 500 and '$ ' not in content and 'took ' not in _content_lower and ('"error":' in _content_head or 'error:' in _content_head or 'err!' in _content_head or 'fatal:' in _content_head or 'exception:' in _content_head or 'traceback' in _content_head or 'command not found' in _content_head or 'invalid syntax' in _content_head or 'failed to' in _content_head) %}
+            {%- set ns2.consecutive_failures = ns2.consecutive_failures + 1 %}
+        {%- else %}
+            {%- set ns2.consecutive_failures = 0 %}
+        {%- endif %}
+        {%- if ns2.prev_role != 'tool' %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {%- if max_tool_response_chars > 0 and content | length > max_tool_response_chars %}
+            {%- set content = content[:max_tool_response_chars] + '\n[TRUNCATED - original length ' ~ (content | length | string) ~ ' chars]' %}
+        {%- endif %}
+        {{- '\n<tool_response>\n' + content }}
+        {%- if ns2.consecutive_failures >= 2 %}
+            {{- '\n\n⚠️ SYSTEM WARNING: ' ~ ns2.consecutive_failures ~ ' consecutive tool errors detected. Your previous approach is incorrect. You MUST use a fundamentally different approach or corrected arguments.' }}
+        {%- elif ns2.consecutive_failures == 1 %}
+            {{- '\n\n⚠️ SYSTEM WARNING: The previous tool call returned an error. Diagnose the failure and retry with completely corrected arguments.' }}
+        {%- endif %}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last %}
+            {{- '<|im_end|>\n' }}
+        {%- else %}
+            {%- set _next_role = _msgs[loop.index0 + 1].role %}
+            {%- if _next_role != 'tool' %}
+                {{- '<|im_end|>\n' }}
+            {%- endif %}
+        {%- endif %}
+    {%- else %}
+        {{- '<|im_start|>user\n[' + message.role + ']: ' + content + '<|im_end|>\n' }}
+    {%- endif %}
+    {%- set ns2.prev_role = message.role %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+    {%- if not ns_state.thinking %}
+        {{- '<think>\n\n</think>\n\n' }}
+    {%- else %}
+        {{- '<think>\n' }}
+    {%- endif %}
+{%- endif %}