You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
description: Adds Claude Sonnet 4.5 1M context for Claude Code, improves Cloud task routing telemetry, and fixes Ollama/LM Studio visibility.
3
+
keywords:
4
+
- roo code 3.28.16
5
+
- new features
6
+
- bug fixes
7
+
image: /img/v3.28.16/v3.28.16.png
8
+
---
9
+
10
+
# Roo Code 3.28.16 Release Notes (2025-10-10)
11
+
12
+
This release adds a 1M‑context variant for Claude Sonnet 4.5 in Claude Code, improves Cloud task routing telemetry, and fixes Ollama/LM Studio provider visibility.
## 1M context for Claude Sonnet 4.5 in Claude Code
17
+
18
+
We've added a 1M‑token context variant of Claude Sonnet 4.5 to Claude Code (thanks ColbySerpa!) ([#8586](https://github.com/RooCodeInc/Roo-Code/pull/8586)):
19
+
20
+
- Fit very large projects and long logs with fewer truncations.
21
+
- Select it via the [1m] suffix when choosing the model; reduces splitting and manual chunking.
22
+
23
+
This makes it easier to work across big repos and lengthy traces without losing context.
24
+
25
+
> **📚 Documentation**: See [Claude Code provider](/providers/claude-code) for model options and context limits.
26
+
27
+
## QOL Improvements
28
+
29
+
* Identify Cloud tasks in the extension bridge to improve diagnostics, logging, and future UI behavior ([#8539](https://github.com/RooCodeInc/Roo-Code/pull/8539))
30
+
* Add the parent task ID in telemetry to improve traceability ([#8532](https://github.com/RooCodeInc/Roo-Code/pull/8532))
31
+
32
+
## Bug Fixes
33
+
34
+
* Make Ollama and LM Studio appear as dynamic providers in provider lists so they can be selected and configured like others ([#8511](https://github.com/RooCodeInc/Roo-Code/pull/8511))
description: Adds Claude Haiku 4.5 model across Anthropic, Bedrock, and Vertex with 200k context and prompt caching, plus a zh‑TW UI label fix.
3
+
keywords:
4
+
- roo code 3.28.17
5
+
- provider updates
6
+
- qol improvements
7
+
---
8
+
9
+
# Roo Code 3.28.17 Release Notes (2025-10-15)
10
+
11
+
This release adds Claude Haiku 4.5 (200k context, prompt caching) and clarifies the zh‑TW “Run command” label.
12
+
13
+
14
+
## QOL Improvements
15
+
16
+
* Clarify the zh‑TW “Run command” label to match the tooltip and reduce confusion (thanks PeterDaveHello!) ([#8631](https://github.com/RooCodeInc/Roo-Code/pull/8631))
17
+
18
+
## Provider Updates
19
+
20
+
* Add Claude Haiku 4.5 across Anthropic, AWS Bedrock, and Vertex AI with 200k context, up to 64k output tokens, images, and prompt caching support ([#8673](https://github.com/RooCodeInc/Roo-Code/pull/8673))
Copy file name to clipboardExpand all lines: docs/update-notes/v3.28.mdx
+6Lines changed: 6 additions & 0 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -100,6 +100,10 @@ This gives teams a higher-capacity OpenAI option without extra configuration.[#8
100
100
***Bedrock 1M Context Checkbox**: Adds checkbox to enable 1M context window support for Claude Sonnet 4 and 4.5 on Bedrock, providing more flexibility for large-scale projects ([#8384](https://github.com/RooCodeInc/Roo-Code/pull/8384))
101
101
* Include reasoning messages in cloud task histories so you can review complete thinking after runs ([#8401](https://github.com/RooCodeInc/Roo-Code/pull/8401))
102
102
103
+
* Identify Cloud tasks in the extension bridge to improve diagnostics, logging, and future UI behavior ([#8539](https://github.com/RooCodeInc/Roo-Code/pull/8539))
104
+
* Add the parent task ID in telemetry to improve traceability ([#8532](https://github.com/RooCodeInc/Roo-Code/pull/8532))
105
+
* Clarify the zh‑TW “Run command” label to match the tooltip and reduce confusion (thanks PeterDaveHello!) ([#8631](https://github.com/RooCodeInc/Roo-Code/pull/8631))
106
+
103
107
## Bug Fixes
104
108
105
109
***AWS Bedrock Claude Sonnet 4.5**: Corrected model identifier for Claude Sonnet 4.5 on AWS Bedrock (thanks sunhyung!) ([#8371](https://github.com/RooCodeInc/Roo-Code/pull/8371))
@@ -158,6 +162,8 @@ This gives teams a higher-capacity OpenAI option without extra configuration.[#8
158
162
***Supernova 1M context**: Upgrades the default Supernova to roo/code-supernova-1-million (1,000,000‑token context). Existing settings auto‑migrate; expect fewer truncations on long tasks. ([#8330](https://github.com/RooCodeInc/Roo-Code/pull/8330))
159
163
***Z.ai GLM‑4.6**: Adds GLM‑4.6 model support with a 200k (204,800) context window and availability across both international and mainland API lines (thanks dmarkey!) ([#8408](https://github.com/RooCodeInc/Roo-Code/pull/8408))
160
164
***Chutes**: Add DeepSeek‑V3.1‑Terminus, DeepSeek‑V3.1‑turbo, DeepSeek‑V3.2‑Exp, GLM‑4.6‑FP8, and Qwen3‑VL‑235B‑A22B‑Thinking to the provider picker ([#8467](https://github.com/RooCodeInc/Roo-Code/pull/8467))
165
+
***Claude Code**: Added a 1M‑context variant of Claude Sonnet 4.5; select via the `[1m]` suffix to fit larger repos and logs with fewer truncations (thanks ColbySerpa!) ([#8586](https://github.com/RooCodeInc/Roo-Code/pull/8586))
166
+
***Claude Haiku 4.5**: Available across Anthropic, AWS Bedrock, and Vertex AI with 200k context, up to 64k output, image input, and prompt caching support ([#8673](https://github.com/RooCodeInc/Roo-Code/pull/8673))
***Roo Code Cloud**: Deprecate “Grok 4 Fast”; show warning if selected and hide when not selected ([#8481](https://github.com/RooCodeInc/Roo-Code/pull/8481))
163
169
***Bedrock**: Claude Sonnet 4 1M‑context handling and compatibility improvements ([#8421](https://github.com/RooCodeInc/Roo-Code/pull/8421))
0 commit comments