This is a list of the top Generative AI tools that we use inside our own business, as well as see in operation across more than 60 companies. The list is of the tools we view as performing the best in this setting, as well as providing the most real value. It is not intended to be an inventory of every tool measured against every criterion, but rather what we see in real-world use across a mixture of small to large businesses in New Zealand.
Notable is the absence of open-source models and tools, of which there are many excellent options available. For most businesses, they lack the expertise and/or resource to implement and manage these themselves.
This list is revised and, where necessary, updated monthly.
This criterion is the most heavily utilised, performing numerous core functions such as evaluation, critical feedback, synthetic research and content ideation. These are also the LLMs are also the workhorse for summarisation, thematic identification, content repurposing and copy preparation.
There are also many specific use-case tools, like Copy.ai and Jasper, however for businesses that have grasped foundational skills, the base models have proven to offer the greatest utility, flexibility and benefit.
All LLMs mentioned here are available on free and paid subscriptions, however, the free use tier for both is severely limited and not practical for meaningful use.
Claude performs significantly better than all other models currently, offering more human-like interactions and outputs. Its ability to reason, interpret images and (for coders) write code then render it live in the window using Artifacts, is exceptional. But Claude has no image generation capability or internet access when compared to ChatGPT 4o.
The crown of best performer seems to shift between Claude and Chat GPT seems to shift with each update, however, Chat GPT makes mention due to its image generation ability and internet access. Chat GPT 4o also allows individuals and businesses to create Custom GPTs, which Claude addresses with Projects, but Custom GPTs are more accessible for many users.
Chat GPT 4o is best suited to business, specifically using the Team Workspace license model. This license model gives businesses centralised control of their subscription costs, access management controls, and the ability to create business-specific custom GPTs. This effective management control, especially the management of IP created via Custom GPTs, is critical in business. When also combined with the added benefit of image generation and internet access, Chat GPT 4o makes the most sense.
There was a clear winner for this category, but many notable mentions for specific use cases, which are typically used on free usage tiers but business.
Hands down the best image generator, especially for photo-realistic output. Midjourney continues to be the gold standard however, like all AI models, suMers from bias – which needs to be actively managed through awareness and training. Important notes: Midjourney-generated images for commercial use require a paid tier license, starting at USD$9. Additionally, Midjourney is accessed via Discord, however after generating 100 images users can use the Midjourney Alpha web interface, significantly increasing productivity.
Flux appeared from nowhere in August and instantly took out the second spot. It is a great all-rounder, with strong prompt adherence and text capabilities, showing a lot of potential. The model is offered via many services, some paid some free.
Ideogram: great at generating text.
Dall-E: good at text, better at illustration style generations, but included with Chat GPT.
Leonardo.AI: excellent niche-specific options, very strong at character reference/ consistency.
This category is fascinating as while still not production quality, some people are doing amazing things. It requires patience and testing, but some are getting great B-roll and short-format video results.
Runway takes out first place for its text-to-video and image-to-video capabilities which while pricey are a cut ahead of the rest. It does require patience to get good results.
Both offer reasonable generation and provide a generous free their, however, both require a lot of patience, as generation times can be quite long. Below is an example we made with Midjourney images, which were then converted using image-to-video in Luma and then a soundtrack generated with Elevenlabs.
This category is specific to images the business already has, which they wish to edit, modify and or otherwise manipulate using Generative AI.
Adobe has incorporated many Generative AI features into Photoshop, which is already established as the gold standard for image editing, making it hard to beat. Already in use by many businesses, it has become the default for serious image editors.
We have not found anyone who has cracked this category. Numerous tools address this category, however none to date have given great performance.
All three generate reasonable base presentation, but with limited direction and/or control, with most requiring paid access after very few uses.
We see this category being used mostly for royalty-free background audio and automated voice generation from existing content. This saves many businesses a lot of money from sound bank subscriptions and voice-over/actor services for lower-value projects. We have also seen some great examples of guerilla-style Social Media campaigns with tremendous success.
Both are currently on the receiving end of legal challenges, but both produce the best music across the board. The genre and vocal styles are impressive and are impossible to distinguish from human-generated music! Most businesses use the free tier, which is also royalty-free, and find they do not reach the volumes requiring a paid license.
Widely regarded as the best in voice generation, translation to other languages retaining the original narrator’s voice, and sound effect generation. ElevenLabs maintain its crown and powers many high-profile initiatives, such as the personalised recaps that were offered during the 2024 Olympics.
This category is specific to editing video or audio podcasts which have already been recorded, not the management of creating them. There were three winners here, each for specific reasons.
Descript allows the user to upload content and then edit it as if editing a Word document, i.e. removing gaps, words and/or sections. It also has a Generative AI feature on paid plans, but the businesses canvassed, all used the free tier which accommodated their monthly requirements.
An entirely free service, offered by Adobe, which does one thing extremely well: improve audio quality. A lot of audio is recorded under less-than-ideal conditions, or on sub-par microphones. Adobe Podcast normalises and corrects audio, making it sound as if professionally recorded in a studio.
Opus Clip excels at taking existing video, intelligently processing the audio and then identifying ‘click-worthy’ short edits for social media. It is very effective at generating Mr Beast styled, captioned, short-format videos and does a great job. Most businesses use the free tier, however, this does limit the amount of video that can uploaded and watermarks the output.
This is a crowded category, with numerous applications including recording and transcription. We have found that one platform stands head and shoulders above the rest for its use of Generative AI.
Like most of the others in this category, Otter can join meetings, has scheduling integration and collaboration tools, but its super-power is insight. Otter sets itself apart by being excellent at post-meeting interrogation and evaluation. You can chat to Otter to ask things like ‘who spoke the most’, ‘how receptive was the client’ and the very powerful ‘how could I have run this meeting better’. Numerous businesses have found this to be of significant value from a staff development and ongoing development perspective.
Peter Mangin is a seasoned technology leader with over 25 years of experience, Peter has a knack for leveraging technology to drive innovation and business transformation. One of his core competencies lies in the practical application of AI, using it as a tool to unlock new possibilities, drive insights, and create value. His current focus is on taking Pure SEO to the next level, embodying his commitment to impactful work.
Located in Auckland, New Zealand, AI Innovisory is your strategic partner in navigating the complex landscape of AI and its transformative impact on businesses. We are not an AI solution provider but a dedicated AI consulting firm that empowers businesses to harness the power of AI in innovative and strategic ways.
If you’re prepared to lead in your industry, we’re here to help you excel. Let’s explore how you can maximise the potential of advanced technologies.
Don’t just be in the game – be ahead of it.