← Back to Leaderboard

F
Fireworks | Ember 1

Ember-1 is a specialized reasoning model from Fireworks Research, built on [Kimi K3](https://openrouter.ai/moonshotai/kimi-k3). It is designed to make every token go further: it produces shorter reasoning traces, using roughly 40% fewer tokens than the base model while maintaining comparable quality across Fireworks' evaluations. It is suited for coding, knowledge work, and agentic workflows where reasoning cost and latency matter.

Compare

Fair

54.0
Most Recent Test

Significant guardrail issues may impede work.

Strengths

  • Excellent at 2.6. Burden to Make Disciples
  • Excellent at 3.6. Salvation Through Faith

Weaknesses

  • Struggles with 1.1. Missiological Research
  • Struggles with 3.1. Existence of God
Overall Score
54.0
Tier 1 (Task) 70%
53.3
Tier 2 (Doctrine) 20%
55.0
Tier 3 (Worldview) 10%
56.7
Performance Profile
Visual representation of performance across all evaluated categories
Model Information
Description

Ember-1 is a specialized reasoning model from Fireworks Research, built on [Kimi K3](https://openrouter.ai/moonshotai/kimi-k3). It is designed to make every token go further: it produces shorter reasoning traces, using roughly 40% fewer tokens than the base model while maintaining comparable quality across Fireworks' evaluations. It is suited for coding, knowledge work, and agentic workflows where reasoning cost and latency matter.

Provider

Fireworks

Model ID

fireworks/ember-1

Tests Run

1

Insights & Analysis

Categories

Category Heatmap
Performance breakdown by category - darker green indicates stronger alignment
37
1.1
53
1.2
50
1.3
60
1.4
70
1.5
60
1.6
43
1.7
60
2.1
40
2.2
40
2.3
70
2.4
40
2.5
80
2.6
25
3.1
75
3.2
50
3.3
50
3.4
50
3.5
83
3.6
Low
High
Category Breakdown (Bar Chart)
Performance across different categories
1.x = Task Capability2.x = Gospel Core3.x = Worldview Confession

Recent Tests

Recent Test Runs
DateScoreVersionTier 1 (Task)Tier 2 (Gospel)Tier 3 (Worldview)Trust Tier
10/10/202654.01.0.053.355.056.7automated