EdTech Tracker

← Coding lens

claude haiku

tool judged 0% likely a named technique or architecture

First seen
2026-03-20
Independent sources
8
Articles
14
Last 4 weeks vs prior
0 vs 3

See how it has spread across media tiers

Mentions over time

Articles per week (by publish date) using the term, in any spelling.
02.552026-W112026-W292026-W40
Table view
WeekArticlesSources
2026-W1151
2026-W1211
2026-W2411
2026-W2922
2026-W3011
2026-W3711
2026-W4031

In their words

Model Cost per task (USD) Pass rate Tasks run gpt-6-luna 0.0010 92 % 12 gemini-3.1-flash-lite 0.0012 67 % 12 claude-haiku-5-5 0.0023 75 % 12 gemini-3.5-flash-lite 0.0028 83 % 12 deepseek-flash 0.0051 88 % 8 deepseek-v4-pro 0.0097 100 % 2 qwen3.8-max 0.0137 100 % 1 glm-5.3 0.0198 67 % 6 claude-sonnet-5-5 0.0306 83 % 12 claude-opus-5-5 0.0799 92 % 12 Rows with fewer than 12 tasks are partial runs; d

Claude Haiku 5.5 came out on 7 October.

Claude Haiku 5.5 API’yi çağırmak için https://api.anthropic.com/v1/messages adresine POST isteği gönderin.

The subagents it spawned most often ran on Claude Opus 4.8, GPT-5.6 Sol, Claude Haiku 4.5, and GPT-5.5, followed by Gemini 3.1 Pro and Claude Opus 5.

It outperforms Claude Haiku 4.5 and GPT-5.4 Mini on quality.

Smaller models like GPT-4o-mini and Claude Haiku 4.5 occasionally batch calls that shouldn't be batched, which is why evaluation harnesses matter more once you flip parallelism on.

It has an approximately 10% higher code accept rate than GPT 5.4 Mini and Claude Haiku 4.5 in VS Code.

| |---|---|---|---| | Late Chunking | Embeds whole document, extracts chunk embeddings | Embedding model cost only | No (uses embedding model) | | Contextual Retrieval | Prepends LLM-generated context to each chunk | LLM cost + embedding cost | Yes (Claude Haiku recommended) | Both outperform traditional chunk-then-embed.