fix(condenser): Retry on empty condensation #1577

csmith49 · 2026-01-02T18:04:20Z

Summary

Adds a graceful failure mechanism when a condenser doesn't forget any events. Instead of throwing an uncaught exception, we now throw a NoCondensationAvailableException which is caught by the base condenser class and returns the view provided.

Notably, this means if we cannot find a condensation (because, e.g., the entirety of the message history is a tool loop) then the condensation will be effectively delayed until a valid chunk of events can be forgotten.

Also updates the condenser integration test to report the number of events condensed in the first condensation, adds unit tests for the exception handling, and updates a handful of other related tests.

Intended to address #1518.

Checklist

If the PR is changing/adding functionality, are there tests to reflect this?
If there is an example, have you run the example to make sure that it works?
If there are instructions on how to run the code, have you followed the instructions and made sure that it works?
If the feature is significant enough to require documentation, is there a PR open on the OpenHands/docs repository with the same branch name?
Is the github CI passing?

Agent Server images for this PR

• GHCR package: https://github.com/OpenHands/agent-sdk/pkgs/container/agent-server

Variants & Base Images

Variant	Architectures	Base Image	Docs / Tags
java	amd64, arm64	`eclipse-temurin:17-jdk`	Link
python	amd64, arm64	`nikolaik/python-nodejs:python3.12-nodejs22`	Link
golang	amd64, arm64	`golang:1.21-bookworm`	Link

Pull (multi-arch manifest)

# Each variant is a multi-arch manifest supporting both amd64 and arm64
docker pull ghcr.io/openhands/agent-server:22eb6f2-python

Run

docker run -it --rm \
  -p 8000:8000 \
  --name agent-server-22eb6f2-python \
  ghcr.io/openhands/agent-server:22eb6f2-python

All tags pushed for this build

ghcr.io/openhands/agent-server:22eb6f2-golang-amd64
ghcr.io/openhands/agent-server:22eb6f2-golang_tag_1.21-bookworm-amd64
ghcr.io/openhands/agent-server:22eb6f2-golang-arm64
ghcr.io/openhands/agent-server:22eb6f2-golang_tag_1.21-bookworm-arm64
ghcr.io/openhands/agent-server:22eb6f2-java-amd64
ghcr.io/openhands/agent-server:22eb6f2-eclipse-temurin_tag_17-jdk-amd64
ghcr.io/openhands/agent-server:22eb6f2-java-arm64
ghcr.io/openhands/agent-server:22eb6f2-eclipse-temurin_tag_17-jdk-arm64
ghcr.io/openhands/agent-server:22eb6f2-python-amd64
ghcr.io/openhands/agent-server:22eb6f2-nikolaik_s_python-nodejs_tag_python3.12-nodejs22-amd64
ghcr.io/openhands/agent-server:22eb6f2-python-arm64
ghcr.io/openhands/agent-server:22eb6f2-nikolaik_s_python-nodejs_tag_python3.12-nodejs22-arm64
ghcr.io/openhands/agent-server:22eb6f2-golang
ghcr.io/openhands/agent-server:22eb6f2-java
ghcr.io/openhands/agent-server:22eb6f2-python

About Multi-Architecture Support

Each variant tag (e.g., 22eb6f2-python) is a multi-arch manifest supporting both amd64 and arm64
Docker automatically pulls the correct architecture for your platform
Individual architecture tags (e.g., 22eb6f2-python-amd64) are also available if needed

openhands-ai · 2026-01-02T18:13:27Z

Looks like there are a few issues preventing this PR from being merged!

GitHub Actions are failing:
- Run tests

If you'd like me to help, just leave a comment, like

@OpenHands please fix the failing actions on PR #1577 at branch `fix/empty-condensation`

Feel free to include any additional details that might help me get this PR into a better state.

_{^{You can manage your notification settings}}

github-actions · 2026-01-02T18:15:53Z

Coverage Report •

File	Stmts	Miss	Cover	Missing
openhands-sdk/openhands/sdk/agent
agent.py	199	71	64%	89, 93, 144, 148–149, 157–159, 169–170, 186–188, 195–197, 199, 203, 206–207, 209–210, 228, 255, 260, 271, 310, 315, 326, 329, 352, 362–363, 384–386, 388, 400–401, 406–407, 427–428, 433, 445–446, 451–452, 490–491, 497–498, 502, 510–511, 519, 522–524, 529–530, 551, 558, 562–563, 601–603, 606–607, 611
openhands-sdk/openhands/sdk/context
view.py	228	111	51%	87, 92, 97–98, 103–104, 109–113, 143–144, 147–153, 156–158, 162, 166–169, 172–173, 179–181, 185–187, 189, 192, 197–201, 204–206, 210–212, 216–219, 222–223, 225, 227, 230–231, 233–234, 236, 240, 242, 244, 247–249, 251, 253–254, 257, 260, 263, 265–266, 268, 284–288, 290, 322–323, 354, 365–366, 374, 377, 433–436, 438–440, 451–452, 454, 456, 478–481, 484, 486–487, 495, 498–499
openhands-sdk/openhands/sdk/context/condenser
base.py	36	11	69%	62, 131–134, 136–137, 139, 143, 147, 151
llm_summarizing_condenser.py	99	64	35%	51, 58, 72, 76–77, 80–83, 86–87, 89, 94, 97–98, 105–107, 113–114, 122, 124–128, 130, 152, 155, 157, 164, 168, 172–176, 178, 201–202, 204, 206–208, 210–212, 214, 218–219, 221–222, 224, 234, 237, 240, 245, 248, 251, 259, 263–264, 270, 272
TOTAL	14574	6930	52%

github-actions · 2026-01-02T18:21:47Z

Hi! I started running the integration tests on your PR. You will receive a comment with the results shortly.

github-actions · 2026-01-02T18:35:38Z

🧪 Integration Tests Results

Overall Success Rate: 92.0%
Total Cost: $2.63
Models Tested: 6
Timestamp: 2026-01-02 18:35:33 UTC

📁 Detailed Logs & Artifacts

Click the links below to access detailed agent/LLM logs showing the complete reasoning process for each model. On the GitHub Actions page, scroll down to the 'Artifacts' section to download the logs.

litellm_proxy_moonshot_kimi_k2_thinking: 📥 View & Download Logs
litellm_proxy_gpt_5.1_codex_max: 📥 View & Download Logs
litellm_proxy_deepseek_deepseek_chat: 📥 View & Download Logs
litellm_proxy_claude_sonnet_4_5_20250929: 📥 View & Download Logs
litellm_proxy_vertex_ai_gemini_3_pro_preview: 📥 View & Download Logs
litellm_proxy_mistral_devstral_2512: 📥 View & Download Logs

📊 Summary

Model	Overall	Integration (Required)	Behavior (Optional)	Tests Passed	Skipped	Total	Cost	Tokens
litellm_proxy_moonshot_kimi_k2_thinking	87.5%	87.5%	N/A	7/8	1	9	$0.78	1,245,526
litellm_proxy_gpt_5.1_codex_max	87.5%	87.5%	N/A	7/8	1	9	$0.16	196,618
litellm_proxy_deepseek_deepseek_chat	100.0%	100.0%	N/A	8/8	1	9	$0.14	1,568,917
litellm_proxy_claude_sonnet_4_5_20250929	100.0%	100.0%	N/A	9/9	0	9	$0.66	597,772
litellm_proxy_vertex_ai_gemini_3_pro_preview	100.0%	100.0%	N/A	9/9	0	9	$0.72	581,227
litellm_proxy_mistral_devstral_2512	75.0%	75.0%	N/A	6/8	1	9	$0.18	427,110

📋 Detailed Results

litellm_proxy_moonshot_kimi_k2_thinking

Overall Success Rate: 87.5% (7/8)
Integration Tests (Required): 87.5% (7/9)
Total Cost: $0.78
Token Usage: prompt: 1,229,776, completion: 15,750, cache_read: 1,094,400
Run Suffix: litellm_proxy_moonshot_kimi_k2_thinking_c160c5c_kimi_k2_run_N9_20260102_182208
Skipped Tests: 1

Skipped Tests:

t08_image_file_viewing: This test requires a vision-capable LLM model. Please use a model that supports image input.

Failed Tests:

t06_github_pr_browsing ⚠️ REQUIRED: No final answer found from agent. Events: 84, LLM messages: 1 (Cost: $0.57)

litellm_proxy_gpt_5.1_codex_max

Overall Success Rate: 87.5% (7/8)
Integration Tests (Required): 87.5% (7/9)
Total Cost: $0.16
Token Usage: prompt: 192,391, completion: 4,227, cache_read: 112,512, reasoning: 2,304
Run Suffix: litellm_proxy_gpt_5.1_codex_max_c160c5c_gpt51_codex_run_N9_20260102_182214
Skipped Tests: 1

Skipped Tests:

t09_token_condenser: This test stresses long repetitive tool loops to trigger token-based condensation. GPT-5.1 Codex Max often declines such requests for efficiency/safety reasons.

Failed Tests:

t06_github_pr_browsing ⚠️ REQUIRED: Agent's final answer does not contain the expected information about the PR content. Final answer preview: I don’t have direct access to external sites (including GitHub) from this environment, so I can’t open that PR to see what’s going on or what @asadm suggested. If you can paste the PR description/comm... (Cost: $0.0063)

litellm_proxy_deepseek_deepseek_chat

Overall Success Rate: 100.0% (8/8)
Integration Tests (Required): 100.0% (8/9)
Total Cost: $0.14
Token Usage: prompt: 1,554,917, completion: 14,000, cache_read: 1,458,496
Run Suffix: litellm_proxy_deepseek_deepseek_chat_c160c5c_deepseek_run_N9_20260102_182214
Skipped Tests: 1

Skipped Tests:

t08_image_file_viewing: This test requires a vision-capable LLM model. Please use a model that supports image input.

litellm_proxy_claude_sonnet_4_5_20250929

Overall Success Rate: 100.0% (9/9)
Integration Tests (Required): 100.0% (9/9)
Total Cost: $0.66
Token Usage: prompt: 584,179, completion: 13,593, cache_read: 502,615, cache_write: 80,546, reasoning: 4,138
Run Suffix: litellm_proxy_claude_sonnet_4_5_20250929_c160c5c_sonnet_run_N9_20260102_182216

litellm_proxy_vertex_ai_gemini_3_pro_preview

Overall Success Rate: 100.0% (9/9)
Integration Tests (Required): 100.0% (9/9)
Total Cost: $0.72
Token Usage: prompt: 558,862, completion: 22,365, cache_read: 371,594, reasoning: 15,607
Run Suffix: litellm_proxy_vertex_ai_gemini_3_pro_preview_c160c5c_gemini_3_pro_run_N9_20260102_182210

litellm_proxy_mistral_devstral_2512

Overall Success Rate: 75.0% (6/8)
Integration Tests (Required): 75.0% (6/9)
Total Cost: $0.18
Token Usage: prompt: 424,390, completion: 2,720
Run Suffix: litellm_proxy_mistral_devstral_2512_c160c5c_devstral_2512_run_N9_20260102_182210
Skipped Tests: 1

Skipped Tests:

t08_image_file_viewing: This test requires a vision-capable LLM model. Please use a model that supports image input.

Failed Tests:

t09_token_condenser ⚠️ REQUIRED: Condensation not triggered. Token counting may not work. (Cost: $0.002)
t02_add_bash_hello ⚠️ REQUIRED: Shell script is not executable (Cost: $0.01)

github-actions · 2026-01-02T18:47:02Z

Hi! I started running the integration tests on your PR. You will receive a comment with the results shortly.

github-actions · 2026-01-02T18:55:28Z

🧪 Integration Tests Results

Overall Success Rate: 98.0%
Total Cost: $2.11
Models Tested: 6
Timestamp: 2026-01-02 18:55:22 UTC

📁 Detailed Logs & Artifacts

Click the links below to access detailed agent/LLM logs showing the complete reasoning process for each model. On the GitHub Actions page, scroll down to the 'Artifacts' section to download the logs.

litellm_proxy_deepseek_deepseek_chat: 📥 View & Download Logs
litellm_proxy_claude_sonnet_4_5_20250929: 📥 View & Download Logs
litellm_proxy_vertex_ai_gemini_3_pro_preview: 📥 View & Download Logs
litellm_proxy_mistral_devstral_2512: 📥 View & Download Logs
litellm_proxy_moonshot_kimi_k2_thinking: 📥 View & Download Logs
litellm_proxy_gpt_5.1_codex_max: 📥 View & Download Logs

📊 Summary

Model	Overall	Integration (Required)	Behavior (Optional)	Tests Passed	Skipped	Total	Cost	Tokens
litellm_proxy_deepseek_deepseek_chat	100.0%	100.0%	N/A	8/8	1	9	$0.05	450,871
litellm_proxy_claude_sonnet_4_5_20250929	100.0%	100.0%	N/A	9/9	0	9	$0.63	514,643
litellm_proxy_vertex_ai_gemini_3_pro_preview	100.0%	100.0%	N/A	9/9	0	9	$0.56	337,211
litellm_proxy_mistral_devstral_2512	87.5%	87.5%	N/A	7/8	1	9	$0.32	772,317
litellm_proxy_moonshot_kimi_k2_thinking	100.0%	100.0%	N/A	8/8	1	9	$0.36	563,530
litellm_proxy_gpt_5.1_codex_max	100.0%	100.0%	N/A	8/8	1	9	$0.19	218,548

📋 Detailed Results

litellm_proxy_deepseek_deepseek_chat

Overall Success Rate: 100.0% (8/8)
Integration Tests (Required): 100.0% (8/9)
Total Cost: $0.05
Token Usage: prompt: 439,097, completion: 11,774, cache_read: 410,816
Run Suffix: litellm_proxy_deepseek_deepseek_chat_d74c039_deepseek_run_N9_20260102_184719
Skipped Tests: 1

Skipped Tests:

t08_image_file_viewing: This test requires a vision-capable LLM model. Please use a model that supports image input.

litellm_proxy_claude_sonnet_4_5_20250929

Overall Success Rate: 100.0% (9/9)
Integration Tests (Required): 100.0% (9/9)
Total Cost: $0.63
Token Usage: prompt: 503,187, completion: 11,456, cache_read: 414,192, cache_write: 88,126, reasoning: 3,115
Run Suffix: litellm_proxy_claude_sonnet_4_5_20250929_d74c039_sonnet_run_N9_20260102_184724

litellm_proxy_vertex_ai_gemini_3_pro_preview

Overall Success Rate: 100.0% (9/9)
Integration Tests (Required): 100.0% (9/9)
Total Cost: $0.56
Token Usage: prompt: 316,425, completion: 20,786, cache_read: 176,898, reasoning: 14,821
Run Suffix: litellm_proxy_vertex_ai_gemini_3_pro_preview_d74c039_gemini_3_pro_run_N9_20260102_184724

litellm_proxy_mistral_devstral_2512

Overall Success Rate: 87.5% (7/8)
Integration Tests (Required): 87.5% (7/9)
Total Cost: $0.32
Token Usage: prompt: 767,170, completion: 5,147
Run Suffix: litellm_proxy_mistral_devstral_2512_d74c039_devstral_2512_run_N9_20260102_184724
Skipped Tests: 1

Skipped Tests:

t08_image_file_viewing: This test requires a vision-capable LLM model. Please use a model that supports image input.

Failed Tests:

t02_add_bash_hello ⚠️ REQUIRED: Shell script is not executable (Cost: $0.01)

litellm_proxy_moonshot_kimi_k2_thinking

Overall Success Rate: 100.0% (8/8)
Integration Tests (Required): 100.0% (8/9)
Total Cost: $0.36
Token Usage: prompt: 551,483, completion: 12,047, cache_read: 488,109
Run Suffix: litellm_proxy_moonshot_kimi_k2_thinking_d74c039_kimi_k2_run_N9_20260102_184724
Skipped Tests: 1

Skipped Tests:

t08_image_file_viewing: This test requires a vision-capable LLM model. Please use a model that supports image input.

litellm_proxy_gpt_5.1_codex_max

Overall Success Rate: 100.0% (8/8)
Integration Tests (Required): 100.0% (8/9)
Total Cost: $0.19
Token Usage: prompt: 214,141, completion: 4,407, cache_read: 108,288, reasoning: 2,240
Run Suffix: litellm_proxy_gpt_5.1_codex_max_d74c039_gpt51_codex_run_N9_20260102_184720
Skipped Tests: 1

Skipped Tests:

t09_token_condenser: This test stresses long repetitive tool loops to trigger token-based condensation. GPT-5.1 Codex Max often declines such requests for efficiency/safety reasons.

csmith49 · 2026-01-02T19:07:01Z

There are two integration test runs here -- the first one has the token condenser test failing for Devstral, but it's the same failure mode as with GPT 5.1. I'm hesitant to mark it skipped because the failure seems to be intermittent.

csmith49 · 2026-01-02T19:55:20Z

IMPORTANT NOTE: This PR is not a complete solution. There are still some strange cases where we can get caught in loops. For example, if a context window exception is thrown we'll end up retrying into infinity.

xingyaoww · 2026-01-02T22:31:34Z

For example, if a context window exception is thrown we'll end up retrying into infinity.

Could we fail hard in this case? IMO fail hard is better here than going into infinite loop though

csmith49 · 2026-01-02T22:37:10Z

Could we fail hard in this case? IMO fail hard is better here than going into infinite loop though

That's doable. The challenge right now is just figuring out when we're in that situation. I've got some draft changes that make that easy though.

all-hands-bot · 2026-01-05T12:19:19Z

[Automatic Post]: I have assigned @xingyaoww as a reviewer based on git blame information. Thanks in advance for the help!

openhands-sdk/openhands/sdk/context/condenser/base.py

xingyaoww

Just one comment, otherwise LGTM

…oft reqs

csmith49 · 2026-01-05T18:33:59Z

Hey @xingyaoww can you take another look now? Here are the changes from what you saw:

The CondensationRequest event has no more enum -- they're all the same, regardless of whether they come from the agent or from the user (via LocalConversation.condense, for example).
The RollingCondenser base class has had the API slightly changed. The function should_condense is replaced by condensation_requirement, which has return type CondensationRequirement | None. The requirements can be either hard or soft and the condenser gets to pick when overriding this function.
- If the requirement is hard and there's no possible condensation, throw an exception. This is the behavior before this change.
- If the requirement is soft and there's no possible condensation, just skip it for now.

This flow lets us wait until the manipulation indices are good when we cross the event count threshold but throw an error in pretty much all other cases.

Next step (in a separate PR) is instead condensing everything if we get a hard requirement and no condensation.

xingyaoww

LGTM! Let's run integration test - good to merge when it passes

github-actions · 2026-01-05T18:40:03Z

Hi! I started running the integration tests on your PR. You will receive a comment with the results shortly.

github-actions · 2026-01-05T18:50:00Z

🧪 Integration Tests Results

Overall Success Rate: 98.0%
Total Cost: $2.06
Models Tested: 6
Timestamp: 2026-01-05 18:49:55 UTC

📁 Detailed Logs & Artifacts

Click the links below to access detailed agent/LLM logs showing the complete reasoning process for each model. On the GitHub Actions page, scroll down to the 'Artifacts' section to download the logs.

litellm_proxy_gpt_5.1_codex_max: 📥 View & Download Logs
litellm_proxy_deepseek_deepseek_chat: 📥 View & Download Logs
litellm_proxy_claude_sonnet_4_5_20250929: 📥 View & Download Logs
litellm_proxy_mistral_devstral_2512: 📥 View & Download Logs
litellm_proxy_moonshot_kimi_k2_thinking: 📥 View & Download Logs
litellm_proxy_vertex_ai_gemini_3_pro_preview: 📥 View & Download Logs

📊 Summary

Model	Overall	Integration (Required)	Behavior (Optional)	Tests Passed	Skipped	Total	Cost	Tokens
litellm_proxy_gpt_5.1_codex_max	100.0%	100.0%	N/A	8/8	1	9	$0.30	540,357
litellm_proxy_deepseek_deepseek_chat	100.0%	100.0%	N/A	8/8	1	9	$0.05	371,158
litellm_proxy_claude_sonnet_4_5_20250929	100.0%	100.0%	N/A	9/9	0	9	$0.63	546,416
litellm_proxy_mistral_devstral_2512	87.5%	87.5%	N/A	7/8	1	9	$0.25	616,679
litellm_proxy_moonshot_kimi_k2_thinking	100.0%	100.0%	N/A	8/8	1	9	$0.20	298,232
litellm_proxy_vertex_ai_gemini_3_pro_preview	100.0%	100.0%	N/A	9/9	0	9	$0.62	388,775

📋 Detailed Results

litellm_proxy_gpt_5.1_codex_max

Overall Success Rate: 100.0% (8/8)
Integration Tests (Required): 100.0% (8/9)
Total Cost: $0.30
Token Usage: prompt: 531,629, completion: 8,728, cache_read: 398,720, reasoning: 5,568
Run Suffix: litellm_proxy_gpt_5.1_codex_max_05f62aa_gpt51_codex_run_N9_20260105_184043
Skipped Tests: 1

Skipped Tests:

t09_token_condenser: This test stresses long repetitive tool loops to trigger token-based condensation. GPT-5.1 Codex Max often declines such requests for efficiency/safety reasons.

litellm_proxy_deepseek_deepseek_chat

Overall Success Rate: 100.0% (8/8)
Integration Tests (Required): 100.0% (8/9)
Total Cost: $0.05
Token Usage: prompt: 357,869, completion: 13,289, cache_read: 330,048
Run Suffix: litellm_proxy_deepseek_deepseek_chat_05f62aa_deepseek_run_N9_20260105_184040
Skipped Tests: 1

Skipped Tests:

t08_image_file_viewing: This test requires a vision-capable LLM model. Please use a model that supports image input.

litellm_proxy_claude_sonnet_4_5_20250929

Overall Success Rate: 100.0% (9/9)
Integration Tests (Required): 100.0% (9/9)
Total Cost: $0.63
Token Usage: prompt: 534,503, completion: 11,913, cache_read: 448,959, cache_write: 84,622, reasoning: 3,243
Run Suffix: litellm_proxy_claude_sonnet_4_5_20250929_05f62aa_sonnet_run_N9_20260105_184048

litellm_proxy_mistral_devstral_2512

Overall Success Rate: 87.5% (7/8)
Integration Tests (Required): 87.5% (7/9)
Total Cost: $0.25
Token Usage: prompt: 611,781, completion: 4,898
Run Suffix: litellm_proxy_mistral_devstral_2512_05f62aa_devstral_2512_run_N9_20260105_184042
Skipped Tests: 1

Skipped Tests:

t08_image_file_viewing: This test requires a vision-capable LLM model. Please use a model that supports image input.

Failed Tests:

t02_add_bash_hello ⚠️ REQUIRED: Shell script is not executable (Cost: $0.0085)

litellm_proxy_moonshot_kimi_k2_thinking

Overall Success Rate: 100.0% (8/8)
Integration Tests (Required): 100.0% (8/9)
Total Cost: $0.20
Token Usage: prompt: 286,290, completion: 11,942, cache_read: 223,488
Run Suffix: litellm_proxy_moonshot_kimi_k2_thinking_05f62aa_kimi_k2_run_N9_20260105_184039
Skipped Tests: 1

Skipped Tests:

t08_image_file_viewing: This test requires a vision-capable LLM model. Please use a model that supports image input.

litellm_proxy_vertex_ai_gemini_3_pro_preview

Overall Success Rate: 100.0% (9/9)
Integration Tests (Required): 100.0% (9/9)
Total Cost: $0.62
Token Usage: prompt: 364,108, completion: 24,667, cache_read: 222,021, reasoning: 18,899
Run Suffix: litellm_proxy_vertex_ai_gemini_3_pro_preview_05f62aa_gemini_3_pro_run_N9_20260105_184038

xingyaoww

LGTM!

Calvin Smith added 8 commits January 2, 2026 10:49

exception for no condensation available

494a11b

fix imports

fbb91c4

adding tests for rolling condenser base class

8a073f8

moving excedption behavior, changing base exception class

dafde2c

linting

6846132

linting

6b2e9d9

yet more linting

f57bc81

updating tests that reference old exception logic

c160c5c

csmith49 added the integration-test Runs the integration tests and comments the results label Jan 2, 2026

minor integration test logging changed

d74c039

csmith49 added integration-test Runs the integration tests and comments the results and removed integration-test Runs the integration tests and comments the results labels Jan 2, 2026

csmith49 marked this pull request as ready for review January 2, 2026 19:05

csmith49 removed the integration-test Runs the integration tests and comments the results label Jan 2, 2026

Calvin Smith and others added 5 commits January 2, 2026 16:55

reasons for a condensation request

053e4c0

view extracts unhandled condensation request reasons

042cbbd

rolling condenser only continues on certain condensation request reasons

af64f9b

linting

40f634f

Merge branch 'main' into fix/empty-condensation

17d7243

all-hands-bot requested a review from xingyaoww January 5, 2026 12:19

Merge branch 'main' into fix/empty-condensation

7169405

xingyaoww reviewed Jan 5, 2026

View reviewed changes

openhands-sdk/openhands/sdk/context/condenser/base.py Outdated Show resolved Hide resolved

xingyaoww reviewed Jan 5, 2026

View reviewed changes

Calvin Smith added 6 commits January 5, 2026 10:08

reverting condensation reasons

b4783da

should_condense -> condensation_requirement

aad0bdf

linting

0efffd9

llm summarizing condenser requirements logic updated to handle hard/s…

d292185

…oft reqs

tests

965e139

linting

27f0bb9

csmith49 requested a review from xingyaoww January 5, 2026 18:34

Merge branch 'main' into fix/empty-condensation

05f62aa

xingyaoww reviewed Jan 5, 2026

View reviewed changes

xingyaoww added the integration-test Runs the integration tests and comments the results label Jan 5, 2026

xingyaoww approved these changes Jan 5, 2026

View reviewed changes

xingyaoww merged commit 1fbf867 into main Jan 5, 2026
35 checks passed

xingyaoww deleted the fix/empty-condensation branch January 5, 2026 18:52

fix(condenser): Retry on empty condensation #1577

fix(condenser): Retry on empty condensation #1577

Uh oh!

Conversation

csmith49 commented Jan 2, 2026 • edited by github-actions bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Checklist

Uh oh!

openhands-ai bot commented Jan 2, 2026

Uh oh!

github-actions bot commented Jan 2, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

github-actions bot commented Jan 2, 2026

Uh oh!

github-actions bot commented Jan 2, 2026

🧪 Integration Tests Results

📁 Detailed Logs & Artifacts

📊 Summary

📋 Detailed Results

litellm_proxy_moonshot_kimi_k2_thinking

litellm_proxy_gpt_5.1_codex_max

litellm_proxy_deepseek_deepseek_chat

litellm_proxy_claude_sonnet_4_5_20250929

litellm_proxy_vertex_ai_gemini_3_pro_preview

litellm_proxy_mistral_devstral_2512

Uh oh!

github-actions bot commented Jan 2, 2026

Uh oh!

github-actions bot commented Jan 2, 2026

🧪 Integration Tests Results

📁 Detailed Logs & Artifacts

📊 Summary

📋 Detailed Results

litellm_proxy_deepseek_deepseek_chat

litellm_proxy_claude_sonnet_4_5_20250929

litellm_proxy_vertex_ai_gemini_3_pro_preview

litellm_proxy_mistral_devstral_2512

litellm_proxy_moonshot_kimi_k2_thinking

litellm_proxy_gpt_5.1_codex_max

Uh oh!

csmith49 commented Jan 2, 2026

Uh oh!

csmith49 commented Jan 2, 2026

Uh oh!

xingyaoww commented Jan 2, 2026

Uh oh!

csmith49 commented Jan 2, 2026

Uh oh!

all-hands-bot commented Jan 5, 2026

Uh oh!

Uh oh!

xingyaoww left a comment

Choose a reason for hiding this comment

Uh oh!

csmith49 commented Jan 5, 2026

Uh oh!

xingyaoww left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

github-actions bot commented Jan 5, 2026

Uh oh!

github-actions bot commented Jan 5, 2026

🧪 Integration Tests Results

📁 Detailed Logs & Artifacts

📊 Summary

📋 Detailed Results

litellm_proxy_gpt_5.1_codex_max

litellm_proxy_deepseek_deepseek_chat

litellm_proxy_claude_sonnet_4_5_20250929

litellm_proxy_mistral_devstral_2512

litellm_proxy_moonshot_kimi_k2_thinking

litellm_proxy_vertex_ai_gemini_3_pro_preview

Uh oh!

xingyaoww left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

csmith49 commented Jan 2, 2026 •

edited by github-actions bot

Loading

github-actions bot commented Jan 2, 2026 •

edited

Loading

xingyaoww left a comment •

edited

Loading