Features & Tools

Complete guide to ConvoVerse capabilities - response tiers, 31 AI models, tools, and credits.

Response Tiers

Choose the right reasoning strategy for your task - from quick answers to deep, verified analysis.

Quick

Fast single-pass response. Cheapest and quickest option.

Speed
Cost
Quality
Best for
Simple questions, quick lookups
Fast answers
Basic tool use
Lowest credit usage

Balanced

Plans before answering. Creates a step-by-step approach, then executes.

Speed
Cost
Quality
Best for
Multi-step tasks, research
Step-by-step planning
Clearer reasoning
Moderate cost

Deep

Thorough analysis with cross-verification by a second AI model.

Speed
Cost
Quality
Best for
Complex research, accuracy-critical tasks
Thorough analysis
Cross-verification
Highest accuracy

LLM Providers

31 models across 10 providers. Switch between them anytime, even mid-conversation.

OpenAI

Full support
GPT-5.5, GPT-5.4, GPT-5.4 Mini, GPT-5.4 Nano
Streaming
Tools
Vision

Anthropic

Full support
Claude Opus 4.8, Sonnet 4.6, Haiku 4.5
Streaming
Tools
Long context

Google

Full support
Gemini 3.1 Pro, Gemini 3.5 Flash, Gemini 3 Flash, Gemini 3.1 Flash-Lite
Streaming
Tools
Fast

xAI

Full support
Grok 4.3
Streaming
Tools
Reasoning

Groq

Ultra-fast
GPT-OSS 120B/20B, Llama 3.3 70B, Llama 3.1 8B - ultra-fast inference
Streaming
Very low latency

Cohere

Full support
Command A (+ Reasoning, Translate), Command R+, R, R 7B
Streaming
RAG-optimized

DeepSeek

Affordable
DeepSeek V4 Pro, V4 Flash
Streaming
Ultra-low cost
Reasoning

Mistral

Full support
Mistral Large 3, Medium 3.5 - open-weight
Streaming
256K context
Multimodal

MiMo

Full support
MiMo V2.5 Pro, V2.5 - 1M context, deep thinking
Streaming
Reasoning
1M context

Alibaba

Full support
Qwen3.7 Max, Qwen3.6 Plus, Qwen3.6 Flash - 1M context, optional thinking
Streaming
Tool use
Reasoning

Model Showdown

Pit two AI models against each other on the same topic - then publish the result and let anyone fork it.

Step 1

Pick two models

Choose Model A and Model B from any of the available providers.

Step 2

Set the intent

Debate, red-team, compare, brainstorm, or open-ended.

Step 3

Referee the rounds

Watch them exchange turns; add a steering note anytime.

Step 4

Publish & fork

Share the debate, or let anyone fork it and continue.

Choose how the models engage

Debate

The two models argue opposing sides.

Red-team

One model stress-tests the other’s reasoning.

Compare

Each answers, then contrasts the two takes.

Brainstorm

They riff and build on each other’s ideas.

Open

Free-form conversation, no set agenda.

Share it

  • • Publish a Showdown to the public feed
  • • Cards show the “Model A vs Model B” matchup
  • • Share a link with a rich preview image

Fork it

  • • Fork any public Showdown into your workspace
  • • Keep the matchup or swap in your own model
  • • Continue the debate from where it left off

Showdowns run in turn batches with a credit cap you set, so you stay in control of cost. Each turn is billed by the model that produced it.

AI Tools

6 built-in tools to extend AI capabilities. Toggle them on or off per conversation.

Knowledge Base

Search your uploaded documents with AI-powered retrieval. Organize files in collections (My Files) and scope which collections the AI searches.

  • Collections (My Files)
  • Hybrid BM25 + vector search
  • Source citations
  • Agentic RAG
  • Google Document AI OCR

Web Search

Search the web for real-time information.

  • Live web results
  • Source links
  • Automatic summarization

URL Fetch

Fetch and analyze content from any URL.

  • HTML parsing
  • Content extraction
  • PDF support

Calculator

Perform mathematical calculations accurately.

  • Basic arithmetic
  • Complex expressions
  • Scientific functions

File Read

Parse and read uploaded file contents.

  • PDF, DOCX, TXT, CSV
  • JSON, YAML, Markdown
  • Structured extraction

Code Execution

Run JavaScript code in a secure sandbox.

  • Sandboxed environment
  • Instant results
  • No network access

Import & Fork Conversations

Import

  • One-click browser extension for ChatGPT, Claude, Grok
  • ChatGPT JSON export file import
  • Preserves message structure and metadata
  • Continue any imported conversation

Fork

  • Fork any public conversation
  • Continue with your own prompts
  • Attribution to original author
  • Fork tree visualization

Credits & Pricing

Subscribe monthly for the best price per credit, or buy one-time top-ups that never expire.

Monthly plans - best price per credit

Lite
9/mo
70 credits / month
€0.13/credit
No rollover
Popular
Pro
19/mo
200 credits / month
€0.10/credit
Rolls over up to 1 month
Power
49/mo
600 credits / month
€0.08/credit
Rolls over up to 1 month

No subscription required - you still get free signup credits to start, and you can cancel anytime from the billing portal.

One-time top-ups - never expire

10 credits
€2.00
€0.20/credit
Popular
50 credits
€9.00
€0.18/credit
100 credits
€15.00
€0.15/credit
500 credits
€70.00
€0.14/credit

How credits work

  • • Plan credits are spent first and renew each cycle (Pro & Power roll over up to 1 month); top-up credits are used next and never expire.
  • • See the exact credit cost before you send - and your balance after.
  • • All prices in EUR, incl. VAT. Secure payments by Stripe, cancel anytime.

Credit Usage Examples

  • • GPT-5.4 / GPT-5.4 Mini: ~0.5-2 credits per message
  • • Claude Sonnet 4.6: ~0.3-1.5 credits per message
  • • GPT-5.4 Nano: ~0.1-0.5 credits per message
  • • DeepSeek Chat: ultra-low cost per message

Cost-Saving Tips

  • • Use Quick tier for simple questions
  • • Choose smaller models for everyday tasks
  • • Subscribe to a monthly plan for the best per-credit rate
  • • Free signup credits to get started

Hybrid Search

Find conversations and documents with our advanced hybrid search system.

BM25

Traditional keyword search with TF-IDF ranking. Fast and accurate for exact matches.

Vector Search

Semantic search using embeddings. Finds conceptually similar content.

Neural Reranking

AI-powered reranking of results for best relevance. Optional but recommended.