[GH-ISSUE #15260] think=false breaks format (structured output) for gemma4 — format constraint silently ignored #71819

New Issue

GiteaMirror · 2026-05-05T02:37:31-05:00

GiteaMirror commented

2026-05-05 02:37:31 -05:00

Originally created by @AIVTDevPKevin on GitHub (Apr 3, 2026).
Original GitHub issue: https://github.com/ollama/ollama/issues/15260

Originally assigned to: @ParthSareen on GitHub.

What is the issue?

Description

When using gemma4:26b-a4b-it-q4_K_M with the format parameter (JSON schema structured output), setting think=false causes the format constraint to be completely ignored. The model outputs plain text instead of the requested JSON structure.

If think is omitted (not sent at all), the format works correctly — but the model then defaults to thinking mode, adding unwanted latency.

This is the same class of bug as #14645 (qwen3.5 series), but confirmed to also affect gemma4. gemma4 uses <|think|> tokens in its chat template for thinking control, similar to how qwen3.5 models handle thinking.

Environment

Ollama version: 0.20.0
Model: gemma4:26b-a4b-it-q4_K_M (SHA: 7121486771cb)
OS: Windows 11 (10.0.26200)
GPU: NVIDIA GeForce RTX 4090 (Driver 582.32)
CPU: Intel Core i9-14900
Tested via: Direct HTTP API calls (curl / requests.post) — not SDK-specific

Minimal Reproduction (via HTTP)

# ❌ FAIL: think=false + format → format is silently IGNORED, outputs plain text
curl -s http://localhost:11434/api/chat -d '{
  "model": "gemma4:26b-a4b-it-q4_K_M",
  "messages": [
    {"role": "system", "content": "You are a helpful assistant."},
    {"role": "user", "content": "Hello!"}
  ],
  "stream": false,
  "think": false,
  "format": {
    "type": "object",
    "properties": {
      "emotion": {"type": "string", "enum": ["happy","sad","neutral"]},
      "response_text": {"type": "string"}
    },
    "required": ["emotion", "response_text"]
  }
}' | python -m json.tool
# → message.content = plain text (NOT JSON), format completely ignored

# ✅ OK: think omitted + format → format works, but model defaults to thinking (extra latency)
curl -s http://localhost:11434/api/chat -d '{
  "model": "gemma4:26b-a4b-it-q4_K_M",
  "messages": [
    {"role": "system", "content": "You are a helpful assistant."},
    {"role": "user", "content": "Hello!"}
  ],
  "stream": false,
  "format": {
    "type": "object",
    "properties": {
      "emotion": {"type": "string", "enum": ["happy","sad","neutral"]},
      "response_text": {"type": "string"}
    },
    "required": ["emotion", "response_text"]
  }
}' | python -m json.tool
# → message.content = valid JSON: {"emotion": "happy", "response_text": "..."}

Test Results (4 scenarios, all via HTTP)

#	Mode	`think`	`format`	Result
1	non-stream	`false`	✅ JSON schema	❌ Plain text — format ignored
2	non-stream	(omitted)	✅ JSON schema	✅ Valid JSON (`emotion=happy`)
3	stream	`false`	✅ JSON schema	❌ Plain text — format ignored
4	stream	(omitted)	✅ JSON schema	✅ Valid JSON (`emotion=happy`)

Expected Behavior

think=false + format should produce valid JSON matching the schema (same as when think is omitted, but without the thinking overhead).

Actual Behavior

When think=false is sent, the format constraint is silently dropped. The model generates unconstrained plain text as if format was never specified.

Root Cause Analysis

Same as described in #14645: Ollama appears to defer format probability masking until it sees the end-of-thinking token. When think=false is set, the thinking tags are closed in the template and the model never outputs the end-of-thinking token, so the masking is never applied.

Notes

Not SDK-specific: Tested with both ollama Python SDK and raw HTTP POST to /api/chat — identical behavior.
Not model-specific to qwen3.5: This affects gemma4 as well. Other models without thinking templates (e.g., gpt-oss:20b) work correctly with think=false + format.
Related: #14645 (qwen3.5), #14850 (qwen3.5:27b, closed as dup), #10929 (invalid JSON with think=true), #10538 (feature request for thinking + structured output)

Relevant log output

Server log shows no errors — all 4 requests returned HTTP 200:

[GIN] 2026/04/03 - 14:27:27 | 200 | 787.3221ms | 127.0.0.1 | POST "/api/chat"  ← Test 1 (think=false, ~0.8s, format IGNORED)
[GIN] 2026/04/03 - 14:27:33 | 200 | 3.8770242s | 127.0.0.1 | POST "/api/chat"  ← Test 2 (think omitted -> default: true, ~3.9s, format OK)
[GIN] 2026/04/03 - 14:27:36 | 200 | 645.5903ms | 127.0.0.1 | POST "/api/chat"  ← Test 3 (think=false, ~0.6s, format IGNORED)
[GIN] 2026/04/03 - 14:27:42 | 200 | 4.0513075s | 127.0.0.1 | POST "/api/chat"  ← Test 4 (think omitted -> default: true, ~4.1s, format OK)

Note: think=false requests complete much faster (~0.7s vs ~4s) because the model skips
thinking — but format constraint is silently ignored, producing plain text instead of JSON.
No warnings or errors are logged server-side when format is ignored.

OS

Windows

GPU

Nvidia

CPU

Intel

Ollama version

0.20.0

Originally created by @AIVTDevPKevin on GitHub (Apr 3, 2026). Original GitHub issue: https://github.com/ollama/ollama/issues/15260 Originally assigned to: @ParthSareen on GitHub. ### What is the issue? ### Description When using `gemma4:26b-a4b-it-q4_K_M` with the `format` parameter (JSON schema structured output), setting `think=false` causes the `format` constraint to be **completely ignored**. The model outputs plain text instead of the requested JSON structure. If `think` is **omitted** (not sent at all), the `format` works correctly — but the model then defaults to thinking mode, adding unwanted latency. This is the same class of bug as #14645 (qwen3.5 series), but confirmed to also affect **gemma4**. gemma4 uses `<|think|>` tokens in its chat template for thinking control, similar to how qwen3.5 models handle thinking. ### Environment - **Ollama version**: 0.20.0 - **Model**: `gemma4:26b-a4b-it-q4_K_M` (SHA: `7121486771cb`) - **OS**: Windows 11 (10.0.26200) - **GPU**: NVIDIA GeForce RTX 4090 (Driver 582.32) - **CPU**: Intel Core i9-14900 - **Tested via**: Direct HTTP API calls (`curl` / `requests.post`) — not SDK-specific ### Minimal Reproduction (via HTTP) ```bash # ❌ FAIL: think=false + format → format is silently IGNORED, outputs plain text curl -s http://localhost:11434/api/chat -d '{ "model": "gemma4:26b-a4b-it-q4_K_M", "messages": [ {"role": "system", "content": "You are a helpful assistant."}, {"role": "user", "content": "Hello!"} ], "stream": false, "think": false, "format": { "type": "object", "properties": { "emotion": {"type": "string", "enum": ["happy","sad","neutral"]}, "response_text": {"type": "string"} }, "required": ["emotion", "response_text"] } }' | python -m json.tool # → message.content = plain text (NOT JSON), format completely ignored # ✅ OK: think omitted + format → format works, but model defaults to thinking (extra latency) curl -s http://localhost:11434/api/chat -d '{ "model": "gemma4:26b-a4b-it-q4_K_M", "messages": [ {"role": "system", "content": "You are a helpful assistant."}, {"role": "user", "content": "Hello!"} ], "stream": false, "format": { "type": "object", "properties": { "emotion": {"type": "string", "enum": ["happy","sad","neutral"]}, "response_text": {"type": "string"} }, "required": ["emotion", "response_text"] } }' | python -m json.tool # → message.content = valid JSON: {"emotion": "happy", "response_text": "..."} ``` ### Test Results (4 scenarios, all via HTTP) | # | Mode | `think` | `format` | Result | |---|------|---------|----------|--------| | 1 | non-stream | `false` | ✅ JSON schema | ❌ **Plain text** — format ignored | | 2 | non-stream | *(omitted)* | ✅ JSON schema | ✅ Valid JSON (`emotion=happy`) | | 3 | stream | `false` | ✅ JSON schema | ❌ **Plain text** — format ignored | | 4 | stream | *(omitted)* | ✅ JSON schema | ✅ Valid JSON (`emotion=happy`) | ### Expected Behavior `think=false` + `format` should produce valid JSON matching the schema (same as when `think` is omitted, but without the thinking overhead). ### Actual Behavior When `think=false` is sent, the `format` constraint is silently dropped. The model generates unconstrained plain text as if `format` was never specified. ### Root Cause Analysis Same as described in #14645: Ollama appears to defer format probability masking until it sees the end-of-thinking token. When `think=false` is set, the thinking tags are closed in the template and the model never outputs the end-of-thinking token, so the masking is **never applied**. ### Notes - **Not SDK-specific**: Tested with both `ollama` Python SDK and raw HTTP `POST` to `/api/chat` — identical behavior. - **Not model-specific to qwen3.5**: This affects `gemma4` as well. Other models without thinking templates (e.g., `gpt-oss:20b`) work correctly with `think=false` + `format`. - Related: #14645 (qwen3.5), #14850 (qwen3.5:27b, closed as dup), #10929 (invalid JSON with think=true), #10538 (feature request for thinking + structured output) ### Relevant log output ```shell Server log shows no errors — all 4 requests returned HTTP 200: [GIN] 2026/04/03 - 14:27:27 | 200 | 787.3221ms | 127.0.0.1 | POST "/api/chat" ← Test 1 (think=false, ~0.8s, format IGNORED) [GIN] 2026/04/03 - 14:27:33 | 200 | 3.8770242s | 127.0.0.1 | POST "/api/chat" ← Test 2 (think omitted -> default: true, ~3.9s, format OK) [GIN] 2026/04/03 - 14:27:36 | 200 | 645.5903ms | 127.0.0.1 | POST "/api/chat" ← Test 3 (think=false, ~0.6s, format IGNORED) [GIN] 2026/04/03 - 14:27:42 | 200 | 4.0513075s | 127.0.0.1 | POST "/api/chat" ← Test 4 (think omitted -> default: true, ~4.1s, format OK) Note: think=false requests complete much faster (~0.7s vs ~4s) because the model skips thinking — but format constraint is silently ignored, producing plain text instead of JSON. No warnings or errors are logged server-side when format is ignored. ``` ### OS Windows ### GPU Nvidia ### CPU Intel ### Ollama version 0.20.0

GiteaMirror added the bug label 2026-05-05 02:37:31 -05:00

GiteaMirror closed this issue

2026-05-05 02:37:33 -05:00

GiteaMirror commented

2026-05-05 02:37:34 -05:00

@johnnyxwan commented on GitHub (Apr 3, 2026):

qwen3.5 series has a similar problem #14645, not sure if they are related

@johnnyxwan commented on GitHub (Apr 3, 2026): qwen3.5 series has a similar problem #14645, not sure if they are related

GiteaMirror commented

2026-05-05 02:37:35 -05:00

@toutjavascript commented on GitHub (Apr 4, 2026):

Thank you for describing this bug. I thought I was going crazy.

@toutjavascript commented on GitHub (Apr 4, 2026): Thank you for describing this bug. I thought I was going crazy.

GiteaMirror commented

2026-05-05 02:37:36 -05:00

@VladimirGav commented on GitHub (Apr 4, 2026):

I have the same problem. JSON breaks when using VladimirGav/gemma4-26b-16GB-VRAM

@VladimirGav commented on GitHub (Apr 4, 2026): I have the same problem. JSON breaks when using VladimirGav/gemma4-26b-16GB-VRAM

GiteaMirror commented

2026-05-05 02:37:37 -05:00

@thiswillbeyourgithub commented on GitHub (Apr 6, 2026):

I think it's really bad that this kind of error is not caught by tests. Structured output is both a major selling point of LLMs and often broken.

@thiswillbeyourgithub commented on GitHub (Apr 6, 2026): I think it's really bad that this kind of error is not caught by tests. Structured output is both a major selling point of LLMs and often broken.

GiteaMirror commented

2026-05-05 02:37:38 -05:00

@MMaturax commented on GitHub (Apr 7, 2026):

In both models 26b and 31b, structured output doesn't work correctly. This isn't related to Ollama; the same problem exists in Google AI Studio. It's impossible to use the Gemma 4 family for structured output and agentic tool calls in this way.

@MMaturax commented on GitHub (Apr 7, 2026): In both models 26b and 31b, structured output doesn't work correctly. This isn't related to Ollama; the same problem exists in Google AI Studio. It's impossible to use the Gemma 4 family for structured output and agentic tool calls in this way.

GiteaMirror commented

2026-05-05 02:37:38 -05:00

@thiswillbeyourgithub commented on GitHub (Apr 7, 2026):

But the model i'm using is gemma4:e4b-it-q8_0

@thiswillbeyourgithub commented on GitHub (Apr 7, 2026): But the model i'm using is gemma4:e4b-it-q8_0

GiteaMirror commented

2026-05-05 02:37:39 -05:00

@AIVTDevPKevin commented on GitHub (Apr 8, 2026):

In both models 26b and 31b, structured output doesn't work correctly. This isn't related to Ollama; the same problem exists in Google AI Studio. It's impossible to use the Gemma 4 family for structured output and agentic tool calls in this way.

Actually, I deployed gemma-4-26B-A4B-it using vLLM to test this out. When interacting with it via the OpenAI-compatible API—strictly enforcing structured outputs while disabling the thinking phase—it successfully returned the correct structured response without any issues.

Given this behavior, it strongly suggests that the root cause still lies within Ollama's implementation rather than an upstream issue with the model itself.

@AIVTDevPKevin commented on GitHub (Apr 8, 2026): > In both models 26b and 31b, structured output doesn't work correctly. This isn't related to Ollama; the same problem exists in Google AI Studio. It's impossible to use the Gemma 4 family for structured output and agentic tool calls in this way. Actually, I deployed [`gemma-4-26B-A4B-it`](https://huggingface.co/google/gemma-4-26B-A4B-it) using vLLM to test this out. When interacting with it via the OpenAI-compatible API—strictly enforcing structured outputs while disabling the thinking phase—it successfully returned the correct structured response without any issues. Given this behavior, it strongly suggests that the root cause still lies within Ollama's implementation rather than an upstream issue with the model itself.

GiteaMirror commented

2026-05-05 02:37:40 -05:00

@johnnyxwan commented on GitHub (Apr 9, 2026):

It has been over a month since the problem is stated in #14645, with at least 2 pull request candidates ( #14660, #14923 ) which can potentially fix this issue. I would appreciate some insight into why this system-breaking, clearly stated, and easily fixable bug has remained unresolved for so long.

@johnnyxwan commented on GitHub (Apr 9, 2026): It has been over a month since the problem is stated in #14645, with at least 2 pull request candidates ( #14660, #14923 ) which can potentially fix this issue. I would appreciate some insight into why this system-breaking, clearly stated, and easily fixable bug has remained unresolved for so long.

GiteaMirror commented

2026-05-05 02:37:41 -05:00

@MingStar commented on GitHub (Apr 12, 2026):

It has been over a month since the problem is stated in #14645, with at least 2 pull request candidates ( #14660, #14923 ) which can potentially fix this issue. I would appreciate some insight into why this system-breaking, clearly stated, and easily fixable bug has remained unresolved for so long.

Was pulling my hair for a day with Ollama v0.20.5.

Could we fix this issue sooner for at least 2 popular model series (qwen and gemma)?

@MingStar commented on GitHub (Apr 12, 2026): > It has been over a month since the problem is stated in [#14645](https://github.com/ollama/ollama/issues/14645), with at least 2 pull request candidates ( [#14660](https://github.com/ollama/ollama/pull/14660), [#14923](https://github.com/ollama/ollama/pull/14923) ) which can potentially fix this issue. I would appreciate some insight into why this system-breaking, clearly stated, and easily fixable bug has remained unresolved for so long. Was pulling my hair for a day with Ollama v0.20.5. Could we fix this issue sooner for at least 2 popular model series (qwen and gemma)?

GiteaMirror commented

2026-05-05 02:37:42 -05:00

@johnnyxwan commented on GitHub (Apr 18, 2026):

The new qwen3.6 is also affected, the problem persists in 0.21.0 as expected.

@johnnyxwan commented on GitHub (Apr 18, 2026): The new qwen3.6 is also affected, the problem persists in 0.21.0 as expected.

GiteaMirror commented

2026-05-05 02:37:43 -05:00

@PureBlissAK commented on GitHub (Apr 18, 2026):

🤖 Automated Triage & Analysis Report

Issue: #15260
Analyzed: 2026-04-18T18:22:49.467732

Analysis

Type: unknown
Severity: medium
Components: unknown

Implementation Plan

Effort: medium
Steps:

This issue has been triaged and marked for implementation.

@PureBlissAK commented on GitHub (Apr 18, 2026):  ## 🤖 Automated Triage & Analysis Report **Issue**: #15260 **Analyzed**: 2026-04-18T18:22:49.467732 ### Analysis - **Type**: unknown - **Severity**: medium - **Components**: unknown ### Implementation Plan - **Effort**: medium - **Steps**: *This issue has been triaged and marked for implementation.*

GiteaMirror commented

2026-05-05 02:37:44 -05:00

@johnnyxwan commented on GitHub (Apr 19, 2026):

@ParthSareen Thank you

@johnnyxwan commented on GitHub (Apr 19, 2026): @ParthSareen Thank you

GiteaMirror commented

2026-05-05 02:37:44 -05:00

@johnnyxwan commented on GitHub (Apr 23, 2026):

The fix is shipped in v0.21.1 and it worked, however, it is scoped only for gemma4. @ParthSareen is there a hope that it can include qwen3.5 and qwen3.6? The very same problem is described in #14645, I am quite sure it can be fixed by just including qwen3.5 and qwen3.6 to the scope. Thank you.

@johnnyxwan commented on GitHub (Apr 23, 2026): The fix is shipped in v0.21.1 and it worked, however, it is scoped only for gemma4. @ParthSareen is there a hope that it can include qwen3.5 and qwen3.6? The very same problem is described in #14645, I am quite sure it can be fixed by just including qwen3.5 and qwen3.6 to the scope. Thank you.

GiteaMirror commented

2026-05-05 02:37:45 -05:00

@Shedletsky commented on GitHub (Apr 23, 2026):

Just came to this thread from Google after wondering if structured output was working with thinking yet for Qwen3:*b in Ollama.

I hacked around this limitation for a project 4 months ago, was hoping to be able to fix it.

It's not clear to me if this is an Ollama issue or a model limitation?

@Shedletsky commented on GitHub (Apr 23, 2026): Just came to this thread from Google after wondering if structured output was working with thinking yet for Qwen3:*b in Ollama. I hacked around this limitation for a project 4 months ago, was hoping to be able to fix it. It's not clear to me if this is an Ollama issue or a model limitation?

GiteaMirror commented

2026-05-05 02:37:46 -05:00

@johnnyxwan commented on GitHub (Apr 24, 2026):

Just came to this thread from Google after wondering if structured output was working with thinking yet for Qwen3:*b in Ollama.

I hacked around this limitation for a project 4 months ago, was hoping to be able to fix it.

It's not clear to me if this is an Ollama issue or a model limitation?

gemma4 is fixed with the merge. For qwen3, models without seperate instruct/thinking such as 8b, latest are also affected. For affected qwen3, or qwen3.5/3.6, please follow #14645.

@johnnyxwan commented on GitHub (Apr 24, 2026): > Just came to this thread from Google after wondering if structured output was working with thinking yet for Qwen3:*b in Ollama. > > I hacked around this limitation for a project 4 months ago, was hoping to be able to fix it. > > It's not clear to me if this is an Ollama issue or a model limitation? gemma4 is fixed with the merge. For qwen3, models without seperate instruct/thinking such as `8b`, `latest` are also affected. For affected qwen3, or qwen3.5/3.6, please follow #14645.

Sign in to join this conversation.

Branches Tags

main

dhiltgen/ci

parth-launch-plan-gating

hoyyeva/anthropic-reference-images-path

parth-anthropic-reference-images-path

brucemacd/download-before-remove

hoyyeva/editor-config-repair

parth-mlx-decode-checkpoints

parth-launch-codex-app

hoyyeva/fix-codex-model-metadata-warning

hoyyeva/qwen

parth/hide-claude-desktop-till-release

hoyyeva/opencode-image-modality

parth-add-claude-code-autoinstall

release_v0.22.0

pdevine/manifest-list

codex/fix-codex-model-metadata-warning

pdevine/addressable-manifest

brucemacd/launch-fetch-reccomended

jmorganca/llama-compat

launch-copilot-cli

hoyyeva/opencode-thinking

release_v0.20.7

parth-auto-save-backup

parth-test

jmorganca/gemma4-audio-replacements

fix-manifest-digest-on-pull

hoyyeva/vscode-improve

brucemacd/install-server-wait

parth/update-claude-docs

brucemac/start-ap-install

pdevine/mlx-update

pdevine/qwen35_vision

drifkin/api-show-fallback

mintlify/image-generation-1773352582

hoyyeva/server-context-length-local-config

jmorganca/faster-reptition-penalties

jmorganca/convert-nemotron

parth-pi-thinking

pdevine/sampling-penalties

jmorganca/fix-create-quantization-memory

dongchen/resumable_transfer_fix

pdevine/sampling-cache-error

jessegross/mlx-usage

hoyyeva/openclaw-config

hoyyeva/app-html

pdevine/qwen3next

brucemacd/sign-sh-install

brucemacd/tui-update

brucemacd/usage-api

jmorganca/launch-empty

fix-app-dist-embed

mxyng/mlx-compile

mxyng/mlx-quant

mxyng/mlx-glm4.7

mxyng/mlx

brucemacd/simplify-model-picker

jmorganca/qwen3-concurrent

fix-glm-4.7-flash-mla-config

drifkin/qwen3-coder-opening-tag

brucemacd/usage-cli

fix-cuda12-fattn-shmem

ollama-imagegen-docs

parth/fix-multiline-inputs

brucemacd/config-docs

mxyng/model-files

mxyng/simple-execute

fix-imagegen-ollama-models

mxyng/async-upload

jmorganca/lazy-no-dtype-changes

imagegen-auto-detect-create

parth/decrease-concurrent-download-hf

fix-mlx-quantize-init

jmorganca/x-cleanup

usage

imagegen-readme

jmorganca/glm-image

mlx-gpu-cd

jmorganca/imagegen-modelfile

parth/agent-skills

parth/agent-allowlist

parth/signed-in-offline

parth/agents

parth/fix-context-chopping

improve-cloud-flow

parth/add-models-websearch

parth/prompt-renderer-mcp

jmorganca/native-settings

jmorganca/download-stream-hash

jmorganca/client2-rebased

brucemacd/oai-chat-req-multipart

jessegross/multi_chunk_reserve

grace/additional-omit-empty

grace/mistral-3-large

mxyng/tokenizer2

mxyng/tokenizer

jessegross/flash

hoyyeva/windows-nacked-app

mxyng/cleanup-attention

grace/deepseek-parser

hoyyeva/remember-unsent-prompt

parth/add-lfs-pointer-error-conversion

parth/olmo2-test2

hoyyeva/ollama-launchagent-plist

nicole/olmo-model

parth/olmo-test

mxyng/remove-embedded

parth/render-template

jmorganca/intellect-3

parth/remove-prealloc-linter

jmorganca/cmd-eval

nicole/nomic-embed-text-fix

mxyng/lint-2

hoyyeva/add-gemini-3-pro-preview

hoyyeva/load-model-list

mxyng/expand-path

mxyng/environ-2

hoyyeva/deeplink-json-encoding

parth/improve-tool-calling-tests

hoyyeva/conversation

hoyyeva/assistant-edit-response

hoyyeva/thinking

origin/brucemacd/invalid-char-i-err

parth/improve-tool-calling

jmorganca/required-omitempty

grace/qwen3-vl-tests

mxyng/iter-client

parth/docs-readme

nicole/embed-test

pdevine/integration-benchstat

parth/remove-generate-cmd

parth/add-toolcall-id

mxyng/server-tests

jmorganca/glm-4.6

jmorganca/gin-h-compat

drifkin/stable-tool-args

pdevine/qwen3-more-thinking

parth/add-websearch-client

nicole/websearch_local

jmorganca/qwen3-coder-updates

grace/deepseek-v3-migration-tests

mxyng/fix-create

jmorganca/cloud-errors

pdevine/parser-tidy

revert-12233-parth/simplify-entrypoints-runner

parth/enable-so-gpt-oss

brucemacd/qwen3vl

jmorganca/readme-simplify

parth/gpt-oss-structured-outputs

revert-12039-jmorganca/tools-braces

mxyng/embeddings

mxyng/gguf

mxyng/benchmark

mxyng/types-null

parth/move-parsing

mxyng/gemma2

jmorganca/docs

mxyng/16-bit

mxyng/create-stdin

pdevine/authorizedkeys

mxyng/quant

parth/opt-in-error-context-window

brucemacd/cache-models

brucemacd/runner-completion

jmorganca/llama-update-6

brucemacd/benchmark-list

brucemacd/partial-read-caps

parth/deepseek-r1-tools

mxyng/omit-array

parth/tool-prefix-temp

brucemacd/runner-test

jmorganca/qwen25vl

brucemacd/model-forward-test-ext

parth/python-function-parsing

jmorganca/cuda-compression-none

drifkin/num-parallel

drifkin/chat-truncation-fix

jmorganca/sync

parth/python-tools-calling

drifkin/array-head-count

brucemacd/create-no-loop

parth/server-enable-content-stream-with-tools

qwen25omni

mxyng/v3

brucemacd/ropeconfig

jmorganca/silence-tokenizer

parth/sample-so-test

parth/sampling-structured-outputs

brucemacd/doc-go-engine

parth/constrained-sampling-json

jmorganca/mistral-wip

brucemacd/mistral-small-convert

parth/sample-unmarshal-json-for-params

brucemacd/jomorganca/mistral

pdevine/bfloat16

jmorganca/mistral

brucemacd/mistral

pdevine/logging

parth/sample-correctness-fix

parth/sample-fix-sorting

jmorgan/sample-fix-sorting-extras

jmorganca/temp-0-images

brucemacd/parallel-embed-models

brucemacd/shim-grammar

jmorganca/fix-gguf-error

bmizerany/nameswork

jmorganca/faster-releases

bmizerany/validatenames

brucemacd/err-no-vocab

brucemacd/rope-config

brucemacd/err-hint

brucemacd/qwen2_5

brucemacd/logprobs

brucemacd/new_runner_graph_bench

progress-flicker

brucemacd/forward-test

brucemacd/go_qwen2

pdevine/gemma2

jmorganca/add-missing-symlink-eval

mxyng/next-debug

parth/set-context-size-openai

brucemacd/next-bpe-bench

brucemacd/next-bpe-test

brucemacd/new_runner_e2e

brucemacd/new_runner_qwen2

pdevine/convert-cohere2

brucemacd/convert-cli

parth/log-probs

mxyng/next-mlx

mxyng/cmd-history

parth/templating

parth/tokenize-detokenize

brucemacd/check-key-register

bmizerany/grammar

jmorganca/vendor-081b29bd

mxyng/func-checks

jmorganca/fix-null-format

parth/fix-default-to-warn-json

jmorganca/qwen2vl

jmorganca/no-concat

parth/cmd-cleanup-SO

brucemacd/check-key-register-structured-err

parth/openai-stream-usage

parth/fix-referencing-so

stream-tools-stop

jmorganca/degin-1

brucemacd/install-path-clean

brucemacd/push-name-validation

brucemacd/browser-key-register

jmorganca/openai-fix-first-message

jmorganca/fix-proxy

jessegross/sample

parth/disallow-streaming-tools

dhiltgen/remove_submodule

jmorganca/ga

jmorganca/mllama

pdevine/newlines

pdevine/geems-2b

jmorganca/llama-bump

mxyng/modelname-7

mxyng/gin-slog

mxyng/modelname-6

jyan/convert-prog

jyan/quant5

paligemma-support

pdevine/import-docs

jmorganca/openai-context

jyan/paligemma

jyan/p2

jyan/palitest

bmizerany/embedspeedup

jmorganca/llama-vit

brucemacd/allow-ollama

royh/ep-methods

royh/whisper

mxyng/api-models

mxyng/fix-memory

jyan/q4_4/8

jyan/ollama-v

royh/stream-tools

roy-embed-parallel

bmizerany/hrm

revert-5963-revert-5924-mxyng/llama3.1-rope

royh/embed-viz

jyan/local2

jyan/auth

jyan/local

jyan/parse-temp

jmorganca/template-mistral

jyan/reord-g

royh-openai-suffixdocs

royh-imgembed

royh-embed-parallel

jyan/quant4

royh-precision

jyan/progress

pdevine/fix-template

jyan/quant3

pdevine/ggla

mxyng/update-registry-domain

jmorganca/ggml-static

mxyng/create-context

jyan/v0.146

mxyng/layers-from-files

build_dist

bmizerany/noseek

royh-ls

royh-name

timeout

mxyng/server-timestamp

bmizerany/nosillyggufslurps

royh-params

jmorganca/llama-cpp-7c26775

royh-openai-delete

royh-show-rigid

jmorganca/enable-fa

jmorganca/no-error-template

jyan/format

royh-testdelete

bmizerany/fastverify

language_support

pdevine/ps-glitches

brucemacd/tokenize

bruce/iq-quants

bmizerany/filepathwithcoloninhost

mxyng/split-bin

bmizerany/client-registry

jmorganca/if-none-match

native

jmorganca/native

jmorganca/batch-embeddings

jmorganca/initcmake

jmorganca/mm

pdevine/showggmlinfo

modenameenforcealphanum

bmizerany/modenameenforcealphanum

jmorganca/done-reason

jmorganca/llama-cpp-8960fe8

ollama.com

bmizerany/filepathnobuild

bmizerany/types/model/defaultfix

rmdisplaylong

nogogen

bmizerany/x

modelfile-readme

bmizerany/replacecolon

jmorganca/limit

jmorganca/execstack

jmorganca/replace-assets

mxyng/tune-concurrency

jmorganca/testing

whitespace-detection

jmorganca/options

upgrade-all

scratch

cuda-search

mattw/airenamer

mattw/allmodelsonhuggingface

mattw/quantcontext

mattw/whatneedstorun

brucemacd/llama-mem-calc

mattw/faq-context

mattw/communitylinks

mattw/noprune

mattw/python-functioncalling

rename

mxyng/install

pulse

remove-first

editor

mattw/selfqueryingretrieval

cgo

mattw/howtoquant

api

matt/streamingapi

format-config

mxyng/extra-args

shell

update-nous-hermes

cp-model

upload-progress

fix-unknown-model

fix-model-names

delete-fix

insecure-registry

ls

deletemodels

progressbar

readme-updates

license-layers

skip-list

list-models

modelpath

matt/examplemodelfiles

distribution

go-opts

1 Participants

Notifications

Due Date

No due date set.

Dependencies

No dependencies set.

Reference: github-starred/ollama#71819

[GH-ISSUE #15260] think=false breaks format (structured output) for gemma4 — format constraint silently ignored #71819

What is the issue?

Description

Environment

Minimal Reproduction (via HTTP)

Test Results (4 scenarios, all via HTTP)

Expected Behavior

Actual Behavior

Root Cause Analysis

Notes

Relevant log output

OS

GPU

CPU

Ollama version

🤖 Automated Triage & Analysis Report

Analysis

Implementation Plan

[GH-ISSUE #15260] `think=false` breaks `format` (structured output) for `gemma4` — format constraint silently ignored #71819