Claude Haiku 5.5

Access Anthropic's Claude Haiku 5.5 through GPTProto for high-volume classification, extraction, routing, support, and subagent work. Test it with one key and a shared balance across available models.

Modalités

Contexte

Exemples d'utilisation de l'API

curl --request POST "https://gptproto.com/v1/chat/completions" \
    --header "Authorization: Bearer $GPTPROTO_API_KEY" \
    --header "Content-Type: application/json" \
    --data '{
        "model": "claude-haiku-5-5",
        "messages": [
            {
                "role": "user",
                "content": "Hello"
            }
        ]
    }'

Tarifs Claude Haiku 5.5

Chat, agents de codage et travail documentaire. Facturé par million de tokens — l'entrée, l'entrée en cache et la sortie sont facturées séparément. GPTProto est 10 % en dessous des tarifs officiels.

Scénario	Liste Claude	OpenRouter	GPTProto	Économies / mois
Personnel10M tokens / mois (4.8M en cache)	$1.37	$1.44	$1.23	−$0.14≈ $1.64 / an
Équipe100M tokens / mois (48M en cache)	$13.68	$14.43	$12.31	−$1.37≈ $16.42 / an
Entreprise500M tokens / mois (240M en cache)	$68.40	$72.16	$61.56	−$6.84≈ $82.08 / an

Coût mensuel par source

OpenRouter

$0.00

Claude

$0.00

GPTProto

$0.00

Économisez $0.00 / mois

Économies annuelles ≈ $0.00 · paiement à l'usage

Claude Haiku 5.5 API

Use Anthropic's fast, small model for frequent developer tasks that need low latency, adjustable reasoning, or long input. Claude Haiku 5.5 combines a 1M-token context window, up to 128K output, text and image input, and adaptive thinking controlled through effort.

1M Context and 128K Output

Process long documents, extended conversations, and large retrieval results within a 1M-token context window. Standard requests can return up to 128K tokens, including thinking tokens.

Adaptive Thinking with Effort

Thinking is adaptive and enabled by default. Use the effort setting to balance reasoning depth, response time, and token consumption; medium is the default on Anthropic's API.

Built for High-Volume Steps

Use Haiku 5.5 for classification, extraction, routing, summaries, database queries, support replies, and focused subagent tasks that run frequently or sit inside a larger workflow.

Text and Image Understanding

Send text or images and receive text output. Use it for document questions, screenshot analysis, chart reading, and visual triage; verify image-input support on the selected GPTProto route.

What Is the Claude Haiku 5.5 API?

Claude Haiku 5.5 is Anthropic's October 7, 2026 update to its small-model line. Anthropic positions it for latency-sensitive and cost-sensitive work such as classification, extraction, routing, live support, browser use, and focused subagent steps. It accepts text and images and returns text.

The model expands the Haiku 4.5 context window from 200K to 1M tokens and its maximum output from 64K to 128K. It is also the first Haiku model with adaptive thinking and adjustable effort. Thinking is on by default, and thinking tokens count toward max_tokens.

Anthropic's direct API uses prompt-length pricing. Requests with up to 100K input tokens use the lower tier; requests over 100K use a higher rate for the full request. This threshold matters when estimating a Claude Haiku 5.5 API cost for long documents or long-running agents. The live GPTProto pricing module is the source of truth for GPTProto billing.

Specification	Claude Haiku 5.5
Provider / release	Anthropic / October 7, 2026
Official Anthropic model ID	`claude-haiku-5-5`
GPTProto model string	`claude-haiku-5-5`
Input → output	Text and images → text
Context window / maximum output	1M tokens / 128K tokens
Thinking / default effort	Adaptive / `medium`
Official direct rate, prompts up to 100K	$0.10 input / $0.50 output per 1M tokens
Official direct rate, prompts over 100K	$0.50 input / $2.50 output per 1M tokens
Reliable knowledge cutoff / training cutoff	June 2026 / June 2026
Open weights	No; API access does not include downloadable weights

Claude Haiku 5.5 API Applications

Classification, extraction, and routing. Label requests, extract fields, route tickets, or choose which model handles the next step. Keep schemas and acceptance checks in the application.

Customer support and live interfaces. Generate short answers, case summaries, query rewrites, and browser assistance. Keep retrieval, permissions, and escalation rules outside the model.

Coding subagents. Assign narrow tasks such as file discovery, change summaries, error triage, or small patches. Use Sonnet 5.5 or Opus 5.5 for more complex, open-ended coding work.

Document and image analysis. Ask questions across long documents or extract data from screenshots and charts. The larger window adds capacity, but citations and validation still matter.

Long tasks and agentic work. Increase effort when a focused step needs more reasoning. Long prompts can cross the 100K pricing threshold, while the newer tokenizer counts the same text as approximately 30% more tokens than Haiku 4.5.

Haiku 4.5 vs Haiku 5.5 vs Sonnet 5.5

Choose by workload rather than family name alone. Haiku 5.5 is the starting point for frequent, focused steps; Sonnet 5.5 remains the stronger option for difficult coding and longer chains of judgment.

Decision factor	Haiku 4.5	Haiku 5.5	Sonnet 5.5
Context / maximum output	200K / 64K	1M / 128K	1M / 128K
Thinking control	Manual thinking budget	Adaptive; default `medium`	Adaptive; default `high`
Official direct input / output rate	$1 / $5 per 1M	$0.10 / $0.50 up to 100K input; $0.50 / $2.50 over 100K	$2 / $10 per 1M
OSWorld 2.1, offline subset	15.7%	72.4%	83.9%
Terminal-Bench 4.0	0.0%	39.2%	70.6%
Best starting point	Existing Haiku 4.5 workloads not yet migrated	High-volume routing, extraction, support, and subagents	Complex coding and multi-step agent work

The benchmark figures above are reported by Anthropic and were not independently reproduced by GPTProto. Run the same prompts, tools, time budget, and acceptance checks on your own workload before moving production traffic.

For GPT 6.1 Sol vs Claude Haiku 5.5, start with Haiku for repeated, well-scoped steps and evaluate GPT 6.1 Sol for demanding coding or agent tasks in an OpenAI-style workflow. Do not compare base token prices alone: include retries, thinking tokens, the 100K Haiku pricing threshold, latency, and accepted-task rate.

Claude Haiku 5.5 API Migration: Checks Before You Switch

Moving from Haiku 4.5 requires more than replacing the model name. Test these changes against the exact endpoint shown in GPTProto's fixed API Usage section:

Confirm the model string. Anthropic uses claude-haiku-5-5 with no date suffix; use the exact ID displayed by GPTProto.
Recount tokens. The same text can produce approximately 30% more tokens than on Haiku 4.5. Recalculate budgets and alerts.
Replace manual thinking budgets. budget_tokens can return a 400 error. Use adaptive thinking and control depth through effort.
Parse blocks by type. A response may begin with a thinking block rather than user-facing text.
Remove custom sampling values. Non-default temperature, top_p, or top_k settings can return a 400 error.
Remove assistant prefills. End messages with a user turn.
Handle new failure states. Add logic for refusal and model_context_window_exceeded, then test before changing the default.

If you are migrating from Anthropic to GPTProto, also verify authentication, endpoint compatibility, streaming events, tool calls, image input, and error objects. Native Anthropic parameters do not necessarily map one-to-one onto every OpenAI-compatible route.

When Should You Choose Claude Haiku 5.5?

Choose Haiku 5.5 for many short or bounded tasks: classification, extraction, routing, retrieval compression, support assistance, visual triage, or focused subagent work. It is also a candidate when Haiku 4.5 lacks enough context or reasoning control.

Choose Claude Sonnet 5.5 for deeper coding judgment or sustained multi-step execution. Keep Claude Haiku 4.5 available until token counts, response parsing, tools, and cost behavior are validated.

Claude Haiku 5.5 API: Common Technical Questions

How much does the Claude Haiku 5.5 API cost on GPTProto?

Use the live pricing module above for the current GPTProto rate. Anthropic's direct list rate is $0.10 input and $0.50 output per 1M tokens up to 100K input, then $0.50 input and $2.50 output when the prompt exceeds 100K.

Is Claude Haiku 5.5 10% off on GPTProto?

GPTProto plans to list Claude Haiku 5.5 at 10% below Anthropic's direct rates. Confirm that the pricing module applies the discount to both the up-to-100K and over-100K tiers before publishing this claim.

How do I get a Claude Haiku 5.5 API key and test the playground?

Create one key in the dashboard, then use the model string and endpoint shown in API Usage. Test a representative prompt in the playground before moving traffic.

What are the Claude Haiku 5.5 context window and output limit?

Anthropic lists a 1M-token context window and up to 128K output for standard requests. Thinking tokens count toward the limit. Do not assume beta batch limits apply on GPTProto unless documented.

Can developers use Claude Haiku 5.5 for coding?

Yes, especially for file discovery, issue triage, summaries, bounded edits, and narrow coding subagents. Compare Sonnet 5.5 for complex repository changes or long-horizon coding.

Is Claude Haiku 5.5 suitable for long tasks and agentic work?

It suits focused agent steps and high-volume subagent work. For long tasks, watch context growth, the 100K pricing threshold, thinking tokens, tool reliability, and early stopping.

Is the Claude Haiku 5.5 API open source?

No. It is a proprietary Anthropic model. GPTProto provides hosted API access, not downloadable weights or self-hosting rights.

Claude Haiku 5.5 vs Sonnet 5.5: which should I choose?

Choose Haiku 5.5 for frequent, bounded steps. Choose Sonnet 5.5 for complex coding, difficult analysis, and longer judgment chains. Measure cost per successful task.

How does Claude Haiku 5.5 compare with GPT 6.1 Sol?

Haiku 5.5 targets high-volume focused work; GPT 6.1 Sol targets demanding coding and agent workflows. Compare both with the same tools, effort budget, latency target, and pass/fail checks.

Articles liés

Guides, comparaisons et mises à jour liés à ce modèle.

Tous les articles

Claude Opus 5.5 vs GPT-6 Sol: Which Is Better for Coding Agents?

A developer-focused comparison of Claude Opus 5.5 and GPT-6 Sol for AI coding assistants and autonomous agents. Explore API pricing, prompt caching, context windows, tool calling, and error recovery—plus what to test before choosing a model for your GPTProto coding workflow.

Claude Opus 5.5 vs Sonnet 5.5: Which Is Worth the Cost for Coding and Agents?

Compare Claude Opus 5.5 and Sonnet 5.5 for coding and agents. See benchmark tradeoffs, cost per task, and GPTProto's 10% lower input/output rates.

What Is Claude Sonnet 5.5? Pricing, Coding Gains, and What Changed

Sonnet 5.5 keeps Sonnet 5’s official token rates. Explore coding results, Claude Code access, upgrade checks, and GPTProto’s 10% off API pricing.