Model techniques map
Taxonomyinference & servingcontext management

taxonomy node · level 2

context management

18 methods filed at this node or below it, from the sources of 12 models.

inference & serving :: context management

Matching aids for the classifier: discard-all context management; keep-recent-k; hierarchical context management; memory compression; context consolidation.

In this branch 18

Everything filed at this node or below it, with one collapsible heading per child node.

filed here 18

Preserved thinking history mode core · 7 sources · 7 quotes
Excluding prior thinking from conversation history default · 6 sources · 6 quotes
Discard-all context management used · 6 sources · 7 quotes
Context compaction used · 3 sources · 3 quotes
Context folding used · 2 sources · 2 quotes
Thinking context management for tool use used · 2 sources · 2 quotes
Context management method used · 1 source · 1 quote
Context management strategy used · 1 source · 1 quote
Discarding tool-call history used · 1 source · 1 quote
Hierarchical context management used · 1 source · 1 quote
Keep-recent-k used · 1 source · 1 quote
Memory compression used · 1 source · 1 quote
Test-time context management for extending token budgets used · 1 source · 1 quote
Preserve thinking optional · 2 sources · 2 quotes
Modality-specific deployment optional · 1 source · 1 quote
Discard-75% evaluated · 2 sources · 2 quotes
Trajectory summarization and rollout re-initiation evaluated · 2 sources · 2 quotes
Summary-based context compression evaluated · 1 source · 1 quote

By model

Which of this branch's techniques each model's own documents describe, and how strongly. Under each model: its strongest treatment anywhere in the branch.

Modeltechniques
NVIDIA-Nemotron-3-Ultra-550B-A55B usedDiscard-all context management usedSummary-based context compression evaluated—
MiMo-V2.5 usedDiscarding tool-call history usedMemory compression used—
GLM-5.3 usedContext management strategy used—
GLM-5.2 usedDiscard-all context management usedHierarchical context management usedKeep-recent-k used—
DeepSeek-V3.2 usedContext management method usedTest-time context management for extending token budgets usedThinking context management for tool use usedDiscard-75% evaluatedDiscard-all context management evaluatedTrajectory summarization and rollout re-initiation evaluated—
Gemma 4 31B defaultExcluding prior thinking from conversation history defaultModality-specific deployment optional—
Kimi K3 corePreserved thinking history mode coreContext compaction used—
Laguna-S-2.1 usedPreserved thinking history mode used—
Qwen3.5-397B-A17B defaultExcluding prior thinking from conversation history defaultContext folding usedDiscard-all context management evaluated—
Qwen3.6-35B-A3B usedPreserved thinking history mode usedPreserve thinking optional—
Qwen3.8-Flash-Next defaultPreserved thinking history mode default—
gpt-oss-120b usedExcluding prior thinking from conversation history used—