# Fireworks AI vs Moonshot AI

> Moonshot built Kimi K3 and serves it at around 33 tokens per second. Fireworks hosts Kimi K3 too, alongside 400+ other models and a fine-tuning stack.

Canonical: https://www.subconscious.dev/compare/fireworks-vs-moonshot-ai · By The Subconscious Team · Updated September 30, 2026

## How they compare

Moonshot AI is the lab, and Kimi K3 is its product: a 2.8 trillion parameter mixture-of-experts model with native vision and 1M context. Vals AI scored it 93.4% on SWE-bench Verified, fourth overall behind closed frontier models. Moonshot's own API charges $3 in and $15 out, with cached input at $0.30. Fireworks lists Kimi K3 among its flagship models, so a team can reach the same weights through a host that also serves DeepSeek V4 Pro and hundreds of others. Fireworks' published speed figures are for DeepSeek V4 Pro, at 167 to 174 tokens per second, so check K3 throughput directly.

Moonshot's first-party API has had growing pains. K3 runs around 33 tokens per second, always thinks, and demand overran capacity days after launch, pausing new subscriptions on July 19. Fireworks adds SOC 2, HIPAA and ISO, marketplace billing and fine-tuning. Moonshot's path suits teams that want the lab's own endpoint, Kimi Code in the terminal, or the cheaper Kimi K2.6 at $0.95 in and $4 out. Any host of K3 should note its custom license, which adds a commercial agreement above $20M in hosting revenue.

## What each one does

### Fireworks AI

Fireworks AI was founded in 2022 by former Meta PyTorch engineers led by CEO Lin Qiao, and it sells speed on open models. Its custom serving stack has posted 167 to 174 tokens per second on DeepSeek V4 Pro in third-party measurements, several times what most GPU peers hit on the same model. The catalog holds 400+ models across text, vision, audio and embeddings, served through an OpenAI-compatible API. In July 2026 it raised a $1.505B Series D at a $17.5B valuation, with a reported $1B+ run rate and 40T+ tokens a day.

### Moonshot AI

Moonshot AI is the Beijing lab behind the Kimi models. Its flagship Kimi K3 launched July 16, 2026 as a 2.8 trillion parameter mixture-of-experts model that activates 16 of 896 experts per token, with native vision and a 1M token context. It is the first open model in the 3T class, and full weights landed on Hugging Face on July 27. The hosted API costs $3 in and $15 out per million tokens, with cached input at $0.30, and it runs through an OpenAI-compatible endpoint, Kimi Code in the terminal, OpenRouter and Cloudflare Workers AI.

## Which is best, and when

### Choose Fireworks AI for

- Running Kimi K3 next to DeepSeek and other models on one API
- Fine-tuning an open model instead of prompting K3
- Compliance needs like SOC 2 and HIPAA

### Choose Moonshot AI for

- Direct access to Kimi K3 and the cheaper K2.6 from the lab
- Terminal coding with Kimi Code
- Repo-scale agents using cached input at $0.30

## At a glance

| Attribute | Fireworks AI | Moonshot AI |
|---|---|---|
| Model access | Open weights | Open weights, custom license |
| Flagship models | DeepSeek V4 Pro, Kimi K3 | Kimi K3, Kimi K2.6 |
| Speed | 167–174 tok/s on DeepSeek V4 Pro | ~33 tok/s on Kimi K3 |
| Price | Fine-tunes served at base price | $3 in, $15 out (Kimi K3) |
| Customization | SFT, DPO, RFT; Training API | Open weights to fine-tune |
| Deployment | Serverless, dedicated GPUs | API, Kimi Code, OpenRouter |
| Long context | Full 1M on DeepSeek V4 Pro | 1M |

## FAQ

### What is the difference between Fireworks AI and Moonshot AI?

Moonshot built Kimi K3 and serves it at around 33 tokens per second. Fireworks hosts Kimi K3 too, alongside 400+ other models and a fine-tuning stack.

### When should I choose Fireworks AI over Moonshot AI?

Running Kimi K3 next to DeepSeek and other models on one API; Fine-tuning an open model instead of prompting K3; Compliance needs like SOC 2 and HIPAA.

### When should I choose Moonshot AI over Fireworks AI?

Direct access to Kimi K3 and the cheaper K2.6 from the lab; Terminal coding with Kimi Code; Repo-scale agents using cached input at $0.30.

### Is Fireworks AI or Moonshot AI cheaper?

Fireworks AI: Fine-tunes served at base price. Moonshot AI: $3 in, $15 out (Kimi K3). The cheaper choice depends on the model and workload.

### Which has more context, Fireworks AI or Moonshot AI?

Fireworks AI: Full 1M on DeepSeek V4 Pro. Moonshot AI: 1M.

## Running long-horizon agents?

If your agents run past 200K tokens, compare both against Subconscious: [Subconscious vs Fireworks AI](https://www.subconscious.dev/compare/subconscious-vs-fireworks.md), [Subconscious vs Moonshot AI](https://www.subconscious.dev/compare/subconscious-vs-moonshot-ai.md).

Full profiles: [Fireworks AI](https://www.subconscious.dev/providers/fireworks.md), [Moonshot AI](https://www.subconscious.dev/providers/moonshot-ai.md).