Fireworks rebuilds Serverless Training API catalog: drops Qwen 3.5 9B, adds DeepSeek V4 Flash and Muse Glimmer 30B
What happened
Between the Aug 31, 2026 and Sep 7, 2026 captures of Fireworks's pricing page, Pulse recorded one change: Serverless Training API model catalog updated: Qwen 3.5 9B (64k, $0.66 prefill) removed and Qwen 3.6 27B replaced by Qwen 3.8 27B at identical pricing; DeepSeek V4 Flash 0731 (262K) and Muse Glimmer 30B (128K) added with per-token prefill/cached/sample/train pricing. Pulse files it as a product change (feature change) with a neutral direction. Pulse's latest capture (Oct 19, 2026) lists On-Demand at $13 per GPU hour (B200 180GB) + $15 per GPU hour (B300 288GB), Serverless at $0.08 per 10M input tokens (embeddings, up to 150M params) + $0.16 per 10M input tokens (embeddings, 150M-350M params) and Fine Tuning at $0.5 per 1M training tokens, LoRA SFT (models up to 16B params).
It lands 6 days after Fireworks's previous change, "Fireworks adds DeepSeek V4 Flash, Muse Glimmer 30B to training API" (Sep 1, 2026). Before that Pulse recorded "Fireworks: On-Demand GPU pricing up 11-30% from Sep 1; region surcharge to 1.5x" (Aug 17, 2026), "Fireworks: New GB300 GPU tier ($18/hr) + region premium pricing" (Aug 11, 2026) and "Fireworks: New Serverless Training API added (Qwen 3.5 9B, 3.6 27B, Kimi K3)" (Aug 5, 2026). Fireworks has been tracked since 2024-11-17.
In the same window Figma, Sketch and Adobe also changed their Marketing & Content pricing pages. Those moves are listed further down, and the Pulse Agent can compare them side by side.
- Serverless · FeatureFeature change
Serverless Training API model catalog updated: Qwen 3.5 9B (64k, $0.66 prefill) removed and Qwen 3.6 27B replaced by Qwen 3.8 27B at identical pricing; DeepSeek V4 Flash 0731 (262K) and Muse Glimmer 30B (128K) added with per-token prefill/cached/sample/train pricing.
| Plan | Field | Before | After | Change |
|---|---|---|---|---|
| Serverless | Feature | Serverless Training API model catalog updated: Qwen 3.5 9B (64k, $0.66 prefill) removed and Qwen 3.6 27B replaced by Qwen 3.8 27B at identical pricing; DeepSeek V4 Flash 0731 (262K) and Muse Glimmer 30B (128K) added with per-token prefill/cached/sample/train pricing. | Feature change | |
Values as written on Fireworks's public pricing page in the two captures Pulse compared; where the page states no number, the cells give the wording before and after.
Evidence
The two captures Pulse comparedFull-page captures of Fireworks's pricing page (fireworks.ai) from Aug 31, 2026 and Sep 7, 2026; the change is marked in the lower part of the page.
Sign in to see the evidence
The two full-page captures with this change marked up are free with an account.
Fireworks, the last 90 days
Full history- Oct 17, 2026improvedFireworks adds Muse Spark Open, DeepSeek V4.1 Flash to Serverless Training
- Sep 28, 2026packagingFireworks adds GLM 5.3 Flash with 200K context, per-token training pricing
- Sep 16, 2026improvedFireworks adds GLM 5.3 to Serverless Training at $14.58/1M train tokens
- Sep 7, 2026improvedFireworks rebuilds Serverless Training API catalog: drops Qwen 3.5 9B, adds DeepSeek V4 Flash and Muse Glimmer 30B · this change
- Sep 1, 2026improvedFireworks adds DeepSeek V4 Flash, Muse Glimmer 30B to training API
- Aug 17, 2026increaseFireworks: On-Demand GPU pricing up 11-30% from Sep 1; region surcharge to 1.5x
- Aug 11, 2026improvedFireworks: New GB300 GPU tier ($18/hr) + region premium pricing
- Aug 5, 2026packagingFireworks: New Serverless Training API added (Qwen 3.5 9B, 3.6 27B, Kimi K3)
- Jul 28, 2026packagingFireworks: new Serverless Training API for pay-per-token LoRA training
- Jul 13, 2026increaseFireworks: On-Demand GPU pricing increased, new B300 tier added
Across Marketing & Content in the same window
All Marketing & Content changesQuestions about this change
Did Fireworks raise prices in September 2026?
No list-price increase was recorded between Aug 31, 2026 and Sep 7, 2026. The change Pulse detected was a product change: feature change.
What changed on Fireworks's pricing page between Aug 31, 2026 and Sep 7, 2026?
Serverless Training API model catalog updated: Qwen 3.5 9B (64k, $0.66 prefill) removed and Qwen 3.6 27B replaced by Qwen 3.8 27B at identical pricing; DeepSeek V4 Flash 0731 (262K) and Muse Glimmer 30B (128K) added with per-token prefill/cached/sample/train pricing. Pulse compared the captures it took on Aug 31, 2026 and Sep 7, 2026; the before/after table above lists each item as written on the page.
How often does Fireworks change its pricing?
Pulse recorded 6 other pricing-page changes for Fireworks in the 90 days before this one; the previous change was "Fireworks adds DeepSeek V4 Flash, Muse Glimmer 30B to training API" on Sep 1, 2026; Fireworks has been tracked since 2024-11-17.