Strixa AI
TopicsSearchPricing
Sign inStart tracking

Signal Detail

Loading signal detail

Reading this signal card's judgment, risks, and evidence.

Strixa AI
TopicsSearchPricing
Sign inStart tracking
Back to Signal Feed
CodeTracked since May 19, 2026

Strip DeepSeek reasoning fields when tools are used on reasoner-bound models

This change fixes a compatibility gap in DeepSeek tool-calling requests by adding auto-detection for `api.deepseek.com` models that route to `deepseek-reasoner` (such as `deepseek-v4-flash` and `deepseek-r1`-style IDs), then enabling a compat flag that removes `reasoning`/`reasoning_effort` whenever `tools` are present in the request, which prevents the reproduced 400 rejection path while keeping `deepseek-v4-pro` behavior unchanged.

DeepSeek APIdeepseek-v4-flashdeepseek-r1deepseek-reasoner

Signal Intelligence

Confidence Level94%
Worth Continued TrackingFinal score 79% with 1 evidence item.
Primary Impact Areas
DeepSeek APIdeepseek-v4-flashdeepseek-r1

What Happened

  • This change fixes a compatibility gap in DeepSeek tool-calling requests by adding auto-detection for `api.deepseek.com` models that route to `deepseek-reasoner` (such as `deepseek-v4-flash` and `deepseek-r1`-style IDs), then enabling a compat flag that removes `reasoning`/`reasoning_effort` whenever `tools` are present in the request, which prevents the reproduced 400 rejection path while keeping `deepseek-v4-pro` behavior unchanged.
  • This change fixes a compatibility gap in DeepSeek tool-calling requests by adding auto-detection for `api.deepseek.com` models that route to `deepseek-reasoner` (such as `deepseek-v4-flash` and `deepseek-r1`-style IDs), then enabling a compat flag that removes `reasoning`/`reasoning_effort` whenever `tools` are present in the request, which prevents the reproduced 400 rejection path while keeping `deepseek-v4-pro` behavior unchanged.
  • 1 evidence item attached for review.

What is Different

Before

Scattered source updates, isolated context, and manual follow-up across multiple feeds.

Now

Introduced a concrete DeepSeek compatibility fix: a new `disableReasoningWhenToolsPresent` flag is auto-enabled for reasoner-bound model IDs on `api.deepseek.com`, and `buildParams` now strips `reasoning` and `reasoning_effort` whenever `tools` exist, with regression tests added to verify the exact keep/drop behavior and opt-out via explicit compat override.

Why Track This

Why It Matters

Tool-using LLM integrations that rely on DeepSeek flash/reasoner models in `oh-my-pi` no longer fail on the first tool-call turn, so operators and developers can keep their subagent workflows running without abrupt 400 errors and without forcing model fallbacks. The guard is implemented in parameter construction after model routing detection, and `deepseek-v4-pro` is explicitly excluded; monitor whether new DeepSeek model IDs or API behavior changes reduce the accuracy of the current ID-based routing heuristic and whether any model marked compatible starts requiring both reasoning and tools together.

Impact

Tool-using LLM integrations that rely on DeepSeek flash/reasoner models in `oh-my-pi` no longer fail on the first tool-call turn, so operators and developers can keep their subagent workflows running without abrupt 400 errors and without forcing model fallbacks. The guard is implemented in parameter construction after model routing detection, and `deepseek-v4-pro` is explicitly excluded; monitor whether new DeepSeek model IDs or API behavior changes reduce the accuracy of the current ID-based routing heuristic and whether any model marked compatible starts requiring both reasoning and tools together.

What To Watch Next

  • Watch whether DeepSeek API becomes a repeated pattern.
  • Track follow-up changes around LLMOps.
  • Compare future signals against this evidence trail.
  • Re-check risk flags: deepseek_model_id_heuristics_miss_new_variants, deepseek_api_contract_shift_on_tool_support.
Open Topic TimelineOpen Technical EventOpen Original Sourcedeepseek_model_id_heuristics_miss_new_variants / deepseek_api_contract_shift_on_tool_support / compat_override_precedence_regression_risk / overly_broad_reasoning_strip_for_edge_models

Supporting Evidence

GITHUB PULL REQUESTHigh Trust

can1357/oh-my-pi PR #1208: fix(ai): drop reasoning_effort when tools present for DeepSeek Flash models

The patch targets the failure where DeepSeek subagent calls with tools returned `400 deepseek-reasoner does not support this tool_choice` by routing a new compatibility rule through `OpenAICompat` and `buildParams` to omit reasoning fields when tool payloads are present.