Fireworks AI Release Notes
23 release notes curated from 1 source by the Releasebot Team. Last updated: Sep 10, 2026
- Sep 9, 2026
- Date parsed from source:Sep 9, 2026
- First seen by Releasebot:Sep 10, 2026
2026-09-09
Fireworks AI adds a training cost estimator to help estimate training job costs before running them.
Training
Training cost estimator
The new training cost estimator helps you estimate what a training job will cost before you run it. Managed and Serverless estimates use published per-token rates. Dedicated estimates use allocated GPU-hour rates. Planning estimates are not quotes.
Original source - Sep 9, 2026
- Date parsed from source:Sep 9, 2026
- First seen by Releasebot:Sep 10, 2026
2026-09-09
Fireworks AI adds a training skill for coding agents to plan runs, estimate cost, and wait for approval before spend.
Training
Training skill for coding agents
A new Fireworks training skill is available for Claude Code, Cursor, Codex, and other compatible coding agents. Describe a training goal in plain language to plan a run, estimate cost, and wait for approval before spend. See Agent Skills for install commands.
Original source All of your release notes in one feed
Join Releasebot and get updates from Fireworks AI and hundreds of other software products.
- Sep 8, 2026
- Date parsed from source:Sep 8, 2026
- First seen by Releasebot:Sep 8, 2026
2026-09-08
Fireworks AI adds platform deployment tags and annotation API changes, with firectl support for setting, unsetting, and listing tags in atomic batches. The REST API now requires custom/ annotation keys for writes and returns only custom/* entries on reads for regular users.
Platform
Deployment tags and annotation API changes
Deployment tags are customer-managed entries stored in a deployment’s annotations map. firectl presents logical keys such as environment; the REST API represents the same key as custom/environment.
firectl: Version 1.8.3 adds deployment tag set, unset, and list, including atomic batch operations.
REST writes: Customer-managed annotation keys must begin with custom/. Bare keys now return HTTP 403 (PERMISSION_DENIED).
REST reads: For regular account users, GetDeployment and ListDeployments return only custom/* annotation entries. Keys outside that namespace are omitted without an error.
Existing stored annotations were not rewritten. Clients using a bare key such as environment should set custom/environment and update reads to use that canonical key.
See Deployment Tags for commands, REST examples, validation rules, and migration guidance.
Original source - Sep 1, 2026
- Date parsed from source:Sep 1, 2026
- First seen by Releasebot:Sep 1, 2026
- Modified by Releasebot:Sep 8, 2026
2026-09-01
Fireworks AI raises Serverless rate limit ceilings by model size, giving smaller models higher limits and larger ones the base ceiling.
Inference
Serverless rate limit ceilings now scale by model size
Serverless adaptive rate limit ceilings now vary by model size tier. Smaller models (< 400B parameters) get higher ceilings; medium models (400B – < 1.6T) get intermediate ceilings; large models (≥ 1.6T) keep the previous base ceilings. See Serverless rate limits for tier thresholds and ceiling values.
Original source - Aug 30, 2026
- Date parsed from source:Aug 30, 2026
- First seen by Releasebot:Aug 31, 2026
2026-08-30
Fireworks AI adds new Serverless Training models DeepSeek V4 Flash 0731, Qwen 3.8 27B, and Muse Glimmer 30B.
New Serverless Training models: DeepSeek V4 Flash 0731, Qwen 3.8 27B, and Muse Glimmer 30B
The following models are now available for LoRA workloads on the shared Serverless Training pool:
- DeepSeek V4 Flash 0731 with up to 262K context
- Qwen 3.8 27B with up to 128K context
- Muse Glimmer 30B with up to 128K context
See the Serverless Training guide for setup and the training model catalog for current availability.
Original source Similar to Fireworks AI with recent updates:
- xAI release notes245 release notes · Latest Sep 11, 2026
- Anthropic release notes818 release notes · Latest Sep 15, 2026
- OpenAI release notes1021 release notes · Latest Sep 14, 2026
- Obsidian release notes113 release notes · Latest Sep 15, 2026
- MiniMax release notes51 release notes · Latest Aug 13, 2026
- Deepseek release notes22 release notes · Latest Sep 10, 2026
- Aug 27, 2026
- Date parsed from source:Aug 27, 2026
- First seen by Releasebot:Aug 31, 2026
- Modified by Releasebot:Sep 8, 2026
2026-08-27
Fireworks AI deprecates several serverless models and recommends migration paths for MiniMax, GPT OSS, Kimi, and DeepSeek.
Inference
Serverless deprecation: MiniMax M2.7, GPT OSS 20B, Kimi K2.6 Turbo/Fast, Kimi K2.7 Code Fast, DeepSeek V4 Pro
The following models are deprecated from serverless effective August 27, 2026.
Recommended migrations:
- MiniMax M2.7 — migrate to MiniMax M3
- GPT OSS 20B — migrate to GPT OSS 120B or Qwen3 8B for lower-latency workloads
- Kimi K2.6 Turbo / Fast — migrate to Kimi K2.6 (standard serving path)
- Kimi K2.7 Code Fast — migrate to Kimi K2.7 Code (standard serving path)
- DeepSeek V4 Pro — migrate to DeepSeek V4 Pro (0813)
- Aug 26, 2026
- Date parsed from source:Aug 26, 2026
- First seen by Releasebot:Aug 28, 2026
2026-08-26
Fireworks AI deprecates Qwen 3.5 9B and Qwen 3.6 27B in Serverless Training and points users to Qwen 3.8 27B.
Serverless Training deprecation: Qwen 3.5 9B and Qwen 3.6 27B
Qwen 3.5 9B and Qwen 3.6 27B are deprecated from Serverless Training effective August 26, 2026. Migrate new and existing Serverless Training workloads to Qwen 3.8 27B. This change applies to the shared Serverless Training pool. Check the training model catalog for availability on other training surfaces.
Original source - Aug 25, 2026
- Date parsed from source:Aug 25, 2026
- First seen by Releasebot:Aug 27, 2026
- Modified by Releasebot:Sep 8, 2026
2026-08-25
Fireworks AI updates SSO docs to support opt-in IdP-initiated SAML login.
Platform
SSO documentation: IdP-initiated SAML
Updated the Custom SSO guide. IdP-initiated SAML is supported as an opt-in (--enable-idp-initiated-sso); the previous troubleshooting copy that said Fireworks only supported SP-initiated login was incorrect.
Original source - Aug 14, 2026
- Date parsed from source:Aug 14, 2026
- First seen by Releasebot:Aug 16, 2026
- Modified by Releasebot:Sep 8, 2026
2026-08-14
Fireworks AI deprecates DeepSeek V4 Flash from serverless and points users to DeepSeek V4 Flash (0731).
Inference
Serverless deprecation: DeepSeek V4 Flash
DeepSeek V4 Flash is deprecated from serverless. Migrate to DeepSeek V4 Flash (0731).
Original source - Jul 16, 2026
- Date parsed from source:Jul 16, 2026
- First seen by Releasebot:Jul 17, 2026
2026-07-16
Fireworks AI adds learning rate scheduler documentation for supervised fine-tuning jobs across firectl and the REST API.
SFT learning rate scheduler documentation
Documented learning rate scheduler settings for supervised fine-tuning jobs, including constant, linear, and cosine schedules via firectl and the REST API lrScheduler object.
Original source - Jun 26, 2026
- Date parsed from source:Jun 26, 2026
- First seen by Releasebot:Jun 21, 2026
- Modified by Releasebot:Sep 8, 2026
2026-06-26
Fireworks AI deprecates Kimi K2.5 and Qwen 3.6 Plus from Serverless, with migrations to newer models.
Inference
Serverless deprecation: Kimi K2.5 and Qwen 3.6 Plus
Serverless deprecation: Kimi K2.5 and Qwen 3.6 Plus are deprecated from serverless.
Recommended migrations:
- Kimi K2.5 — migrate to Kimi K2.6
- Qwen 3.6 Plus — migrate to Qwen 3.7 Plus
- Jun 17, 2026
- Date parsed from source:Jun 17, 2026
- First seen by Releasebot:Jun 21, 2026
- Modified by Releasebot:Sep 8, 2026
2026-06-17
Fireworks AI deprecates MiniMax M2.5 in serverless and directs users to migrate to MiniMax M2.7.
Inference
Serverless deprecation: MiniMax M2.5
MiniMax M2.5 is deprecated from serverless. Migrate to MiniMax M2.7.
Original source - Jun 15, 2026
- Date parsed from source:Jun 15, 2026
- First seen by Releasebot:Jun 21, 2026
- Modified by Releasebot:Sep 8, 2026
2026-06-15
Fireworks AI adds GLM 5.2 to the Model Library for inference.
- Jun 12, 2026
- Date parsed from source:Jun 12, 2026
- First seen by Releasebot:Jun 21, 2026
- Modified by Releasebot:Sep 8, 2026
2026-06-12
Fireworks AI adds Kimi K2.7 Code, MiniMax M3, and Qwen 3.7 Plus to its Model Library for inference.
Inference
New models: Kimi K2.7 Code, MiniMax M3, and Qwen 3.7 Plus
The following models are now available in the Model Library:
- Kimi K2.7 Code
- MiniMax M3
- Qwen 3.7 Plus
- Jun 10, 2026
- Date parsed from source:Jun 10, 2026
- First seen by Releasebot:Jun 21, 2026
2026-06-10
Fireworks AI deprecates audio inference and image generation.
Audio inference and image generation deprecation
Audio inference and image generation are deprecated.
Original source
Curated by the Releasebot team
Releasebot is an aggregator of official release notes from hundreds of software vendors and thousands of sources.
Our editorial process involves the manual review and audit of release notes procured with the help of automated systems.