vs
Cohere vs Boundless
Cohere sells enterprise models, embeddings and reranking for private or on-prem use. Boundless serves open models plus GPT and Claude, most at 1M context.
By The Subconscious Team · Updated
Cohere vs Boundless: key differences
Cohere's Command A lists at $2.50 in and $10 out per million tokens with 256K context, and the open Command A+ has 128K. Embed 5 and Rerank 4 cover retrieval. Models run on Cohere's API, Bedrock, Azure and OCI, on Model Vault instances from $4 an hour, or in a private VPC or on-prem with fine-tuning inside that environment. Boundless serves open models such as DeepSeek V4.1 Flash at $0.20 in and $1.00 out and GLM-5.3 at $1.12 in and $3.52 out, most with 1M context, and resells GPT and Claude. Its tailored tier reserves capacity that Boundless engineers tune and operate, and it offers no fine-tuning.
Cohere is the pick for regulated enterprises that need RAG and agents inside their own network, and for adding embeddings or reranking to search. Boundless fits startups that want long context and a mix of open and closed models through OpenAI or Anthropic SDKs, with zero data retention in the US or Europe.
What Cohere and Boundless do
Cohere
Cohere is a Toronto-based lab that sells models and platforms to banks, governments and large enterprises rather than consumers. Its generative line is the Command family. Command A+, released May 20, 2026, is a 218B-parameter mixture-of-experts model with 25B active, published under Apache 2.0 with a 128K context window, and it combines reasoning, vision, translation and tool use in one set of weights. Command A has a 256K window and lists at $2.50 in and $10 out per million tokens, while Command R7B costs $0.0375 in. June 2026 added North Mini Code, a 30B Apache 2.0 coding model, and the lineup also includes Aya multilingual models and Transcribe for speech.
Example models: Command A+, Command A, Embed 5, Rerank 4
Full Cohere profileBoundless
Boundless launched in September 2026 as an inference partner for AI-native startups. It sells two things. The Model API serves pre-tuned models on shared capacity, billed per token from prepaid credits with a budget per API key. Tailored inference reserves capacity for one customer, and Boundless engineers pick the model on the customer’s own evals, tune precision, batching, caching, speculative decoding and parallelism, then run the deployment against uptime, throughput and latency targets written into the contract. A free two-week evaluation tests those targets on real traffic against the customer’s current provider first.
Example models: GLM-5.3, DeepSeek V4.1 Flash, Kimi K3
Full Boundless profileShould you choose Cohere or Boundless?
Cohere vs Boundless at a glance
| Attribute | Cohere | Boundless |
|---|---|---|
| Model access | Closed, plus open Command A+ | Open weights, plus closed GPT and Claude |
| Flagship models | Command A+, Command A, Embed 5, Rerank 4 | GLM-5.3, DeepSeek V4.1 Flash, Kimi K3 |
| Speed | 375 tok/s on Command A+ W4A4, per Cohere | 342 tok/s on DeepSeek V4.1 Flash 1017 (Vercel) |
| Price | $0.0375–$2.50 in, $0.15–$10 out per 1M | $0.04–$2.30 in, $0.21–$11.40 out per 1M (open) |
| Customization | Enterprise fine-tuning, incl. private | Serving-stack tuning, no fine-tuning |
| Deployment | API, Bedrock, Azure, OCI, VPC, on-prem | Serverless API, reserved tailored capacity |
| Long context | 256K on Command A; 128K on A+ | 1M context on most models |
Frequently asked questions
What is the difference between Cohere and Boundless?
Cohere sells enterprise models, embeddings and reranking for private or on-prem use. Boundless serves open models plus GPT and Claude, most at 1M context.
When should I choose Cohere over Boundless?
On-prem and private VPC deployment; Embed 5 and Rerank 4; Fine-tuning inside your network.
When should I choose Boundless over Cohere?
1M context on most models; GPT and Claude beside open models; Free two-week evaluation.
Is Cohere or Boundless cheaper?
Cohere: $0.0375–$2.50 in, $0.15–$10 out per 1M. Boundless: $0.04–$2.30 in, $0.21–$11.40 out per 1M (open). The cheaper choice depends on the model and workload.
Which has more context, Cohere or Boundless?
Cohere: 256K on Command A; 128K on A+. Boundless: 1M context on most models.
Related comparisons
Running long-horizon agents?
If your agents run past 200K tokens, compare both against Subconscious. Our inference stack treats a long-horizon trace as the primary workload, so speed, cost, and accuracy hold up deep into the trace.