# Cerebras vs Inception

> Both sell speed. Cerebras gets it from a wafer-scale chip running open models, and Inception from diffusion models that generate many tokens at once on standard GPUs.

Canonical: https://www.subconscious.dev/compare/cerebras-vs-inception · By The Subconscious Team · Updated October 11, 2026

## How they compare

Cerebras lists GPT-OSS 120B near 3,000 tokens per second at $0.35 in and $0.75 out, but its shared catalog is just GPT-OSS 120B and Qwen 3.8 27B, and other models need dedicated endpoints or partners. Inception's Mercury 2.5 is a closed reasoning model with 260K context, tool use and JSON mode at $0.20 in and $0.75 out per million. Inception reports over 1,100 tokens per second, and Artificial Analysis measured about 640.

On published numbers Cerebras is faster, and it runs open weights. Inception is cheaper on input, documents a 260K context, and sells through AWS Bedrock and Azure Foundry. Speed matters less when an agent mostly waits on tools. Pick Cerebras for raw speed on GPT-OSS, and Inception for fast reasoning with tool use on a cloud you already buy from.

## What each one does

### Cerebras

Cerebras builds a single chip the size of a silicon wafer, and its inference cloud is the fastest public host on the models it serves. Its developer table lists GPT-OSS 120B near 3,000 tokens per second, about six times Groq on the same weights, at $0.35 in and $0.75 out per million. The public shared catalog is thin, just GPT-OSS 120B and Qwen 3.8 27B since Gemma 4 31B moved to dedicated-only in September 2026. More model families live on dedicated endpoints and through partners like OpenRouter, Hugging Face, Vercel and AWS Marketplace.

### Inception

Inception builds diffusion large language models (dLLMs). Every other LLM on this list generates text one token at a time; Inception's Mercury models start from a rough draft of a whole block and refine it over a few passes, producing many tokens in parallel. That ties speed less to output length and lets the models run fast on ordinary NVIDIA GPUs rather than custom chips. The company was founded by professors from Stanford, UCLA and Cornell behind core work on diffusion, Flash Attention and DPO, and is led by Stanford's Stefano Ermon. It raised a $50M seed in November 2025 led by Menlo Ventures, with Mayfield, M12, NVentures, Snowflake and Databricks, plus Andrew Ng and Andrej Karpathy as angels.

## Which is best, and when

### Choose Cerebras for

- Fastest output on GPT-OSS 120B
- Streaming UIs and live autocomplete
- Open weights at very high speed

### Choose Inception for

- Fast reasoning with tool use and 260K context
- Buying through Bedrock or Azure
- Voice agents with Mercury Voice

## At a glance

| Attribute | Cerebras | Inception |
|---|---|---|
| Model access | Open weights | Closed weights (Mercury dLLMs) |
| Flagship models | GPT-OSS 120B, Qwen 3.8 27B | Mercury 2.5, Mercury 2 |
| Speed | ~3,000 tok/s on GPT-OSS 120B | 1,100+ tok/s (vendor); ~640 measured |
| Price | $0.35 in, $0.75 out (GPT-OSS 120B) | $0.20 in, $0.75 out (Mercury 2.5) |
| Customization | - | Fine-tuning on request |
| Deployment | Shared API, dedicated, partners | Serverless API, Bedrock, Azure, private |
| Long context | 131K on GPT-OSS 120B, 128K on Qwen 3.8 27B | 260K |

## FAQ

### What is the difference between Cerebras and Inception?

Both sell speed. Cerebras gets it from a wafer-scale chip running open models, and Inception from diffusion models that generate many tokens at once on standard GPUs.

### When should I choose Cerebras over Inception?

Fastest output on GPT-OSS 120B; Streaming UIs and live autocomplete; Open weights at very high speed.

### When should I choose Inception over Cerebras?

Fast reasoning with tool use and 260K context; Buying through Bedrock or Azure; Voice agents with Mercury Voice.

### Is Cerebras or Inception cheaper?

Cerebras: $0.35 in, $0.75 out (GPT-OSS 120B). Inception: $0.20 in, $0.75 out (Mercury 2.5). The cheaper choice depends on the model and workload.

### Which has more context, Cerebras or Inception?

Cerebras: 131K on GPT-OSS 120B, 128K on Qwen 3.8 27B. Inception: 260K.

## Running long-horizon agents?

If your agents run past 200K tokens, compare both against Subconscious: [Subconscious vs Cerebras](https://www.subconscious.dev/compare/subconscious-vs-cerebras.md), [Subconscious vs Inception](https://www.subconscious.dev/compare/subconscious-vs-inception.md).

Full profiles: [Cerebras](https://www.subconscious.dev/providers/cerebras.md), [Inception](https://www.subconscious.dev/providers/inception.md).