AD
Track all your projects in one dashboard. Get 📊stats, 🔥heatmaps and 👀recordings in one self-hosted dashboard.
uxwizz.com
#248

Ember-1

Fireworks Release: 2026-09-28 Tested on: 2026-09-28 08:17 fireworks/ember-1::max
~2.8T total (~104B active)MoEClosedEstimated
(max) (high) (low) (none)

Summary

Ember-1 scores 5.9 on AI BENCHY and ranks #248. It has 7.5 reliability, a 57.6% pass rate, $2.792 total cost, and 98.92s average response time.

What makes Ember-1 unique: It stands out most in General Intelligence, where it ranks #1, while Instructions following is its weakest area at #13.

Model facts

Researched on 2026-09-28

Estimated
Parameters
~2.8T total (~104B active)
Architecture
MoE
Availability
Closed
License
-

Best estimate from public evidence; the vendor did not disclose every value. Fireworks reports post-training on Kimi K3 for shorter reasoning. Architecture and parameter estimates inherit Kimi K3's reported 2.8T total and 104B active parameters; Ember-specific counts are not disclosed. The exact Ember checkpoint is offered as a serverless research preview, with no public weights identified.

Parameter source 1 Parameter source 2 Availability source

Score

5.9

Consistency

7.7

Reliability

7.5

Total Cost (Current Price)

$2.792

Total Output Tokens

180,806

Total Input Tokens

26,449

Input Price

$3.000 / 1M

Output Price

$15.000 / 1M

Tests Correct

Wrong Tests: 12

Attempt pass rate: 57.6%

Flaky tests

6

Flaky tests had mixed outcomes across runs (at least one pass and one fail).

Response Time (avg)

98.92s

Response Time (max): 587.08s

Response Time (total): 2176.20s

API error: 8 Wrong answer: 3 No answer: 1

Hamster playing table tennis

Prompt: Create a detailed SVG illustration of a hamster playing table tennis.

#248 Ember-1

max
Cost
$0.163
Time
229.4s
Tokens
11,002 tok

Charts

Choose the first model, then click a second model to open a side-by-side page.

Total Output Tokens

Score vs Total Output Tokens

Quick Compare

Ember-1maxvsDots 3 Note PreviewmediumFree AvailableEmber-1maxvsGLM 5.3 FlashlowEmber-1maxvsStep 3.5 FlashmediumEmber-1maxvsGPT-5.4 MininoneEmber-1maxvsQwen3.5-35B-A3BnoneEmber-1maxvsSeed 2.1 TurbononeEmber-1maxvsQwen3.5-FlashnoneEmber-1maxvsSpace Bunny AlphalowEmber-1maxvsGPT-5 NanomediumEmber-1maxvsGPT-5.4none

Category Breakdown

Category Score Consistency Tests Correct
Anti-AI Tricks 8.7 7.9
Coding 5.2 4.4
Combined 6.5 10.0
Data parsing and extraction 6.4 5.9
Domain specific 3.6 7.2
General Intelligence 10.0 10.0
Instructions following 6.2 5.8
Puzzle Solving 7.7 10.0
Tool Calling 3.0 10.0
Trivia 3.0 10.0

Compared models