AI 模型情報

deepseek/deepseek-ocr-2

novita-ai/ocr-2

出品方： novita-ai · 發布 2026-01-27

⚠ 本模型為社群微調 / 衍生版本，並非廠商官方發布。

$0.030
輸入 / 1M token
$0.030
輸出 / 1M token
8K
上下文長度
8K
最大輸出

Prices in USD per 1M tokens. Unknown means the provider does not publish per-token pricing.

能力清單

✗ 工具呼叫✗ 推理? 結構化輸出✓ 附件✓ 開放權重? 溫度可調
支援模態: 輸入 text, image · 輸出 text

Model fit scores

0–100 · higher is better

These scores reward declared capabilities, context size, price and provider availability — they are not benchmark results. Use them as a directional signal alongside your own evaluation.

Coding1
  • Tool calling0/40
  • Structured output0/20
  • Reasoning0/10
  • Context window (100K → 1M)0/20
  • Provider availability1/10
Agents6
  • Tool calling0/35
  • Structured output0/25
  • Reasoning0/15
  • Output token limit5/15
  • Provider availability1/10
JSON / structured output20
  • Structured output / JSON mode0/50
  • Tool calling0/20
  • Temperature control0/10
  • Price-friendly for high-volume20/20
Cost efficiency91
  • Headline price (log-scaled)91/95
  • Has prompt-cache pricing0/5
Long context0
  • Context ≥ 100K0/100
Vision61
  • Accepts image input50/50
  • Context window (10K → 1M)0/30
  • Has published price10/10
  • Provider availability1/10
Production-readiness43
  • Number of independent providers5/40
  • Has published per-token price20/20
  • Context window ≥ 8K8/15
  • No data inconsistencies across providers10/10
  • Official model (not derivative)0/15

Cost Efficiency Index

Open full calculator →

Estimated cost using the recommended provider's headline rate. Each scenario fixes average input/output tokens — the assumptions are shown in the third column.

ScenarioCostAssumption
RAG answer
per 1,000 RAG answers
$0.17
< $0.01 per request
5K input tokens (query + 4 retrieved chunks of ~1K each) and a 500-token answer. Typical SaaS knowledge-base bot.
Support ticket triage
per 10,000 tickets
$0.33
< $0.01 per request
1K input tokens (ticket body + system prompt) and a 100-token JSON classification reply. High-volume customer support.
Data extraction
per 1,000 documents
$0.07
< $0.01 per request
2K input tokens (a single document page) and a 500-token JSON extraction. ETL / invoice / form pipelines.
Code review
per 1,000 PRs
$0.27
< $0.01 per request
8K input tokens (diff + surrounding files) and a 1K-token review comment. PR-bot workloads.
Agent step
per 1,000 steps
$0.38
< $0.01 per request
12K input tokens (long-running tool history) and a 600-token tool-call decision. Cost per agent step.

定價詳情

推薦定價來自 novita-ai · deepseek/deepseek-ocr-2

$0.030
輸入
$0.030
輸出

於 1 家供應商可用

服務商服務商模型 ID輸入 / 1M輸出 / 1M上下文發布日期
NovitaAI
novita-ai
deepseek/deepseek-ocr-2$0.030$0.0308K2026-01-27

Frequently asked questions

How much does deepseek/deepseek-ocr-2 cost?

deepseek/deepseek-ocr-2 costs $0.030 per 1M input tokens and $0.030 per 1M output tokens, sourced from novita-ai. Cache reads, audio tokens and >200K-context tiers (where applicable) are listed in the Pricing detail block above.

What is the context window of deepseek/deepseek-ocr-2?

deepseek/deepseek-ocr-2 has a context window of 8K tokens, with a max output of 8K tokens per reply. This is the total combined size of prompt + completion.

Does deepseek/deepseek-ocr-2 support tool calling?

No. deepseek/deepseek-ocr-2 does not support tool calling (function calling). If your workflow requires it, look at the /capabilities/tool-calling list for alternatives.

Does deepseek/deepseek-ocr-2 support structured output / JSON mode?

Support for structured output / JSON-schema-constrained decoding is not reported for deepseek/deepseek-ocr-2 in our data source. Verify with novita-ai's official documentation before relying on it in production.

Can deepseek/deepseek-ocr-2 accept image input?

Yes. deepseek/deepseek-ocr-2 accepts both text and image input. Vision pricing per image is usually billed on top of the regular token rate — check novita-ai's docs for the exact rule.

Is deepseek/deepseek-ocr-2 open-weight?

Yes. deepseek/deepseek-ocr-2's weights are publicly available, so you can self-host or fine-tune. Note that open weights ≠ open source — the training data and code are typically not released.

What are the best alternatives to deepseek/deepseek-ocr-2?

If deepseek/deepseek-ocr-2 doesn't fit, consider Ling-2.6-1T, Ling-2.6-flash, PaddleOCR-VL. Each one targets the same use case — see the Related links below for direct head-to-head pages.

Where does this data come from?

All numbers come from the public models.dev API and are normalised into a single canonical model record. We re-pull daily and write any changes (price, context, capability) to the /changelog page.

Explore more

More novita-ai models

Capability lists this model is in

See also

最近更新：

Data is sourced from models.dev and normalized for comparison. Prices and capabilities may change. Always verify critical production decisions with the provider's official documentation.