ellamiELLAMI / CATALOG
Choose a model for text, images or video. Compare capabilities and pricing.
Select a model type to compare prices.
Models: 64
Code, documents and complex tasks in a single workflow.
Model ID: gpt-6.1-sol
Long prompts above 272,000 input tokens: input $4 · output $15 / 1M tokens.
Above 272,000 input tokens, the entire request is charged at 2× input and 1.5× output. Standard prices, excluding Batch, Flex and Fast.
Developer rateDevelopment, codebase work and autonomous agents.
Model ID: gpt-6-sol
Long prompts above 272,000 input tokens: input $4 · output $15 / 1M tokens.
Above 272,000 input tokens, the entire request is charged at 2× input and 1.5× output. Standard prices, excluding Batch, Flex and Fast.
Developer rateDeep analysis, research and demanding engineering tasks.
Model ID: gpt-6-astra
Long prompts above 272,000 input tokens: input $20 · output $75 / 1M tokens.
Above 272,000 input tokens, the entire request is charged at 2× input and 1.5× output. Standard prices, excluding Batch, Flex and Fast.
Developer rateImage generation and editing from prompts and references.
1024×1024 · low · Input billed separately
Model ID: gpt-image-2.5-sunburst
The card shows the minimum image output cost among the listed sizes and qualities. In the token table, input refers to text and output to image tokens. Input images are priced separately. Total cost depends on size, quality and token usage, rather than a fixed price per image.
Developer rateVisual generation, variations and targeted image edits.
1024×1024 · low · Input billed separately
Model ID: gpt-image-2.5-flare
The card shows the minimum image output cost among the listed sizes and qualities. In the token table, input refers to text and output to image tokens. Input images are priced separately. Total cost depends on size, quality and token usage, rather than a fixed price per image.
Developer rateA versatile model for image generation and editing.
1024×1024 · low · Input billed separately
Model ID: gpt-image-2
The card shows the minimum image output cost among the listed sizes and qualities. In the token table, input refers to text and output to image tokens. Input images are priced separately. Total cost depends on size, quality and token usage, rather than a fixed price per image.
Developer rate
Feature development, bug fixing and document work.
Complex code, reviews and long-running agentic tasks.
Demanding reasoning and long-horizon autonomous work.
Complex reasoning, architecture and agentic development.
Code, analysis and everyday professional work.
An established model for programming, writing and analysis.
Multimodal analysis, code and multi-step tasks.
Model ID: gemini-3.8-flash
Regular rates: $1.5 input / $7.5 output. A promotion through December 31, 2026 reduces these to $0.75 / $3.75. The catalog shows regular rates. Reasoning tokens count as output; search is billed separately.
Developer rateFast agents, long documents and multimedia analysis.
Model ID: gemini-3.7-flash
Regular rates: $1.5 input / $7.5 output. A promotion through December 31, 2026 reduces these to $0.75 / $3.75. The catalog shows regular rates. Reasoning tokens count as output; search is billed separately.
Developer rateImage generation and editing with output up to 4K.
0.5K · Input billed separately
Model ID: nano-banana-2
CLODEX uses nano-banana-2; the original Google API ID is gemini-3.1-flash-image. The card shows output cost for one 1K image. Estimated output costs: $0.045 (0.5K), $0.067 (1K), $0.101 (2K), $0.151 (4K); input is charged separately.
Developer rate
Code, reasoning and visual analysis with a large context.
Model ID: deepseek-v4.1-flash
Off-peak: input $0.15 · output $0.6 / 1M tokens.
The catalog shows peak rates. Off-peak prices are 50% lower. Peak hours: Monday–Friday, 01:00–04:00 and 06:00–10:00 UTC.
Developer rateA large model for complex reasoning and development.
Model ID: deepseek-v4-pro
Off-peak: input $0.66 · output $1.98 / 1M tokens.
The catalog shows peak rates. Off-peak prices are 50% lower. Peak hours: Monday–Friday, 01:00–04:00 and 06:00–10:00 UTC.
Developer rate
Development, knowledge work and agents with visual input.
Model ID: grok-4.7
Long prompts above 200,000 input tokens: input $4 · output $12 / 1M tokens.
Base rates for up to 200,000 input tokens. Above 200,000: $4 input / $12 output per 1M. Built-in tools are billed separately.
Developer rate
Multimodal reasoning and code with up to 1M context.
Complex development and long agentic workflows.
Code, visual understanding and long-video analysis.
Model ID: qwen3.8-flash
Regular Singapore / International rates: $0.15 / $0.47. Global regions use different rates: $0.113 / $0.382 per 1M tokens.
Developer rate
Multimodal development and agentic tasks.
Model ID: MiniMax-M3
Long prompts above 512,000 input tokens: input $0.6 · output $2.4 / 1M tokens.
Current Standard rates after the permanent 50% price reduction. For M3 above 512,000 input tokens: $0.6 input / $2.4 output. Priority is priced separately.
Developer rateProgramming and workflow automation.
Professional development, deep analysis and agentic work.
Model ID: gpt-5.6-sol
Long prompts above 272,000 input tokens: input $8 · output $30 / 1M tokens.
Published GPT 5.6 Sol rate: $4 / $20 per 1M; OpenAI identifies it as promotional pricing available at least through November 21, 2026. A later regular rate is not published. Above 272,000 input tokens: $8 / $30 per 1M for the entire request.
Developer rateCode and reasoning with a balance of quality and cost.
Model ID: gpt-5.6-terra
Long prompts above 272,000 input tokens: input $4 · output $18 / 1M tokens.
Above 272,000 input tokens, the entire request is charged at 2× input and 1.5× output. Standard prices, excluding Batch, Flex and Fast.
Developer rateComplex development and long-running tool-based tasks.
Agentic development, architecture and visual tasks.
Deep reasoning, research and autonomous tasks.
Fast multimodal model for code and visual analysis.
Complex reasoning, programming and agent workflows.
Software development and multi-step tool use.
Multimodal reasoning, code and agentic workflows.
Model ID: grok-4.6
Long prompts above 200,000 input tokens: input $4 · output $12 / 1M tokens.
Base rates for up to 200,000 input tokens. Above 200,000: $4 input / $12 output per 1M. Built-in tools are billed separately.
Developer rateApp development and tool-based task automation.
Faster M2.7 variant for development and work agents.
Faster code generation and tool use.
A lightweight model for code, fast responses and high-volume tasks.
Model ID: gpt-6-luna
Long prompts above 272,000 input tokens: input $0.2 · output $0.75 / 1M tokens.
Above 272,000 input tokens, the entire request is charged at 2× input and 1.5× output. Standard prices, excluding Batch, Flex and Fast.
Developer rateA fast, economical variant for everyday work.
Model ID: gpt-5.6-luna
Long prompts above 272,000 input tokens: input $0.4 · output $1.8 / 1M tokens.
Above 272,000 input tokens, the entire request is charged at 2× input and 1.5× output. Standard prices, excluding Batch, Flex and Fast.
Developer rateComplex reasoning, development and tool use.
Model ID: gpt-5.5
Long prompts above 272,000 input tokens: input $10 · output $45 / 1M tokens.
Above 272,000 input tokens, the entire request is charged at 2× input and 1.5× output. Standard prices, excluding Batch, Flex and Fast.
Developer rateCode, analysis and professional tasks with a large context.
Model ID: gpt-5.4
Long prompts above 272,000 input tokens: input $5 · output $22.5 / 1M tokens.
Above 272,000 input tokens, the entire request is charged at 2× input and 1.5× output. Standard prices, excluding Batch, Flex and Fast.
Developer rateA compact model for programming and work agents.
Fast short tasks, classification and data extraction.
Agentic development, bug fixing and repository work.
Model ID: gpt-5.3-codex
OpenAI has announced API retirement on 2027-04-01.
Developer rateAn economical model for code and well-defined tasks.
The lightest GPT 5 variant for brief responses and data processing.
A compact model for code, instructions and long documents.
Fast conversations, text tasks and image analysis.
Complex code, deep analysis and long-running agentic work.
Architecture, programming and tool-based tasks.
App development, analysis and document work.
Fast responses, compact agents and everyday code.
A fast multimodal model for code and analysis.
Model ID: gemini-3.6-flash
Regular rates: $1.5 input / $7.5 output. A promotion through December 31, 2026 reduces these to $0.75 / $3.75. The catalog shows regular rates. Reasoning tokens count as output; search is billed separately.
Developer rateA lightweight multimodal model for fast, high-volume tasks.
Economical responses, data extraction and multimedia processing.
Model ID: gemini-3.1-flash-lite
The $0.25 input rate applies to text, images and video; audio input costs $0.5 per 1M tokens. Reasoning tokens count as output.
Developer rateComplex reasoning, long documents and multimodal analysis.
Model ID: gemini-3.1-pro-preview
Long prompts above 200,000 input tokens: input $4 · output $18 / 1M tokens.
Base rates for up to 200,000 input tokens. Above 200,000: $4 input / $18 output per 1M tokens. Reasoning tokens count as output; search is billed separately.
Developer rateProgramming, knowledge work and agentic tasks.
Model ID: grok-4.5
Long prompts above 200,000 input tokens: input $4 · output $12 / 1M tokens.
Base rates for up to 200,000 input tokens. Above 200,000: $4 input / $12 output per 1M. Built-in tools are billed separately.
Developer rateMultilingual code and multi-step tool use.
A faster M2.1 variant for code and automation.
An economical model for programming and work agents.
Video generation with audio from text and images.
720p / 1080p
Model ID: clodex-video/veo-3.1
Billed per second of successfully generated video with audio. Veo 3.1: $0.4/s for 720p and 1080p, $0.6/s for 4K. Fast: $0.1/s (720p), $0.12/s (1080p), $0.3/s (4K). Lite: $0.05/s (720p), $0.08/s (1080p).
Developer rateFast video generation with audio for iteration and finished content.
720p
Model ID: clodex-video/veo-3.1-fast
Billed per second of successfully generated video with audio. Veo 3.1: $0.4/s for 720p and 1080p, $0.6/s for 4K. Fast: $0.1/s (720p), $0.12/s (1080p), $0.3/s (4K). Lite: $0.05/s (720p), $0.08/s (1080p).
Developer rateVideo generation with audio in 720p and 1080p.
720p
Model ID: clodex-video/veo-3.1-lite
Billed per second of successfully generated video with audio. Veo 3.1: $0.4/s for 720p and 1080p, $0.6/s for 4K. Fast: $0.1/s (720p), $0.12/s (1080p), $0.3/s (4K). Lite: $0.05/s (720p), $0.08/s (1080p).
Developer rateVideo generation from text and multimodal references, up to 2K.
768P
Model ID: clodex-video/minimax-h3
Output: $0.08/s at 768P, $0.13/s at 2K. Audio input is free; the first 5 input images are free, then $0.04 per image. Input video is charged separately at $0.08/s or $0.13/s according to output resolution.
Developer rateVideo generation with audio, with image and audio reference support.
480p
Model ID: grok-imagine-video-1.5
Video output: $0.08/s (480p), $0.14/s (720p), $0.25/s (1080p). Input images are billed separately at $0.01 per image. These are original xAI API rates.
Developer rateExpressive motion video from text and images.
768P · 6s
Model ID: clodex-video/hailuo-2.3
Fixed output price: $0.28 per 6-second 768P video; $0.56 per 10-second video; $0.49 per 6-second 1080P video. The card uses the equivalent per-second price ($0.28 / 6) for comparison; the actual charge is per full clip.
Developer rateNo matching models. Try another name, type or developer.
These are reference model developer rates. Available models and current Ellami prices for your API key: GET /api/v1/models.
Text is priced per 1 million input and output tokens. Images start at the listed price per image. Video is priced per second or clip. Size, quality and billing conditions are under “Model details”.
Create an API key in Console and set base_url / baseURL: https://www.ellami.pro/api/v1. Use a model ID from the public API catalog in an OpenAI-compatible client. Request examples & documentation ↗