FireworksReleased Sep 23, 2026
Proprietary
Ember 1
Strongest in Mathematical Reasoning (#26 of 176) and Programmatic Reasoning (#49 of 177). Five of seven dimensions are estimated, because no results are published yet.
Composite IQ ?
106
Effective cost
$15.87/1M
$3 in · $15 out
Speed
Not measured
No speed data yet
Context
1Mtokens
1,048,576 tokens in
Evidence
2of 7 dims
2 of 39 benchmarks
IQ profile
Seven dimensions, averaged equally into Composite IQ
MeasuredEstimated95% range
#1 in dimensionTop-10 median
IQ dimensions
Ranks are among models scored in each dimension. Ranges are model-based estimates, not calibrated confidence intervals.
DimensionIQPosition · 70–160RankEvidencevs Top 10
Ember 195% range (model-based, uncalibrated)Top-10 median#1 in dimension100 = average human
Benchmark results
2 of 39 Composite IQ benchmarks have a published result for Ember 1. Benchmarks without a result stay listed so gaps are visible.
Abstract Reasoning IQ 89 · estimatedNo published results0/5
No published results yet for Abstract Reasoning. The 89 shown is estimated from Ember 1's other dimensions and will be replaced as results come in. How estimates work
ARC-AGI-3ARC-AGI-2ARC-AGI-1Chess PuzzlesMystery Game Puzzles
Mathematical Reasoning IQ 132 · measuredBest: #22 of 48 on ProofBench 1.11/5
BenchmarkScorePercentileRankLeader · source
ProofBench 1.1
Full source leaderboard reconciliation v1.1 · Oct 27
61.0%
#22 / 48
Leader fable-5.1 · 100%
Full source leaderboard reconciliation v1.1 · Oct 27
Full source leaderboard reconciliation v1.1 · Oct 27
Not yet reported
FrontierMath Tier 4FrontierMath Tier 1-3MathArenaAIME
Academic Reasoning IQ 100 · estimatedNo published results0/6
No published results yet for Academic Reasoning. The 100 shown is estimated from Ember 1's other dimensions and will be replaced as results come in. How estimates work
Humanity's Last ExamGPQA DiamondCritPtSciCodeMMLU-ProMMMU-Pro
Programmatic Reasoning IQ 126 · measuredBest: #30 of 43 on IOI V21/6
BenchmarkScorePercentileRankLeader · source
IOI V2
Full source leaderboard reconciliation v2 · Oct 27
46.3%
#30 / 43
Leader Gemini 4 Argon · 100%
Full source leaderboard reconciliation v2 · Oct 27
Full source leaderboard reconciliation v2 · Oct 27
Not yet reported
LiveCodeBenchTerminal-Bench 4.0ProgramBenchSWE-rebenchMirrorCode
Computer Use IQ 102 · estimatedNo published results0/5
No published results yet for Computer Use. The 102 shown is estimated from Ember 1's other dimensions and will be replaced as results come in. How estimates work
BrowseCompOSWorld-VerifiedMCP AtlasArena.ai Agent ArenaAgents' Last Exam
Reliability IQ 96 · estimatedNo published results0/6
No published results yet for Reliability. The 96 shown is estimated from Ember 1's other dimensions and will be replaced as results come in. How estimates work
SimpleQA VerifiedaaOmniscienceBullshitBench v2IFBenchAA Long Context Reasoning v1.1FACTS Grounding
Knowledge Work IQ 95 · estimatedNo published results0/6
No published results yet for Knowledge Work. The 95 shown is estimated from Ember 1's other dimensions and will be replaced as results come in. How estimates work
AA-Briefcase v1.1GDPval-AA v2.1AutomationBench-AAGDP.pdfAA-AnalystAgentEnterpriseOps-Gym-AA
Release & availability
Facts from the publisher's announcement and documentation.
- Released
- Developer
- Fireworks
- Access
- Available through Fireworks API and OpenRouter.
- Weights
- ProprietaryHosted access only
- Inputs
- textimage
- Context
- 1,048,576 tokens
- API price
- $3 in · $15 outper 1M tokens · full pricing
What Fireworks says
- Fireworks post-trained Kimi K3 to shorten reasoning traces.
- The publisher reports about 40% fewer tokens; this is a provider claim, not an imported independent score.
Publisher statements, not AI IQ measurements.
About this score
How much of Ember 1's IQ is measured, and how the rest is estimated.
2/ 39benchmark results
2/ 7dimensions measured
5estimated
0/5Abstract
1/5Math
0/6Academic
1/6Coding
0/5Computer Use
0/6Reliability
0/6Knowledge Work
Bar height = share of benchmarks with resultsEstimatedNo score
Measured results are shown as published
Every score in the benchmark tables comes from its source unchanged. Composite IQ averages all seven dimensions equally.
Estimates fill the gaps
Abstract Reasoning, Academic Reasoning, Computer Use, Reliability and Knowledge Work have no published results yet, so their IQ is estimated from Ember 1's other dimensions.
A missing result is not a zero
Benchmark versions and evaluation harnesses can differ between sources, so small gaps between models may not be meaningful.
Compare Ember 1
Side-by-side pages for the models it is most often weighed against.
#1 overall
gpt-6.1-sol
IQ141 35 higherEffective cost$6.30 2.5× cheaper
US (Other) #1
Beam
IQ114 8 higher
Closest IQ
gpt-5.4-nano
IQ106 same IQEffective cost$1.63 9.7× cheaper
Most similar profile
Hy4 Preview
IQ112 6 higherEffective cost$3.30 4.8× cheaper
Similar price
gemini-3-pro
IQ114 8 higherEffective cost$14.78 about the same
#2 overall
opus-5.5
IQ141 35 higherEffective cost$35.73 2.3× pricier
Ember 1Comparison model
Frequently asked questions
Quick answers about Ember 1, drawn from the same data as the rest of this page.
How AI IQ scores models →Get this data from the API →- What is the IQ of Ember 1?
- Ember 1 has an estimated IQ of 106, which ranks #93 out of 176 public models on the AI IQ leaderboard. The score combines seven equally weighted reasoning dimensions using measured benchmark results and conservative estimates for missing coverage.
- What is Ember 1 best at?
- Among the dimensions with published benchmark results, Ember 1 ranks highest in Mathematical Reasoning (IQ 132, #26 of 176) and Programmatic Reasoning (IQ 126, #49 of 177).
- What is the EQ of Ember 1?
- Ember 1 doesn't have an EQ score yet. EQ (emotional reasoning) is calculated from EQ-Bench 4 and AttuneBench, and neither has published a result for Ember 1. AI IQ doesn't estimate EQ from other benchmarks, so the score appears once one of those results is published.
- Which benchmarks has Ember 1 been tested on?
- Ember 1 has published results on 2 of the 39 benchmarks behind AI IQ's Composite IQ, including ProofBench 1.1 and IOI V2.
- How much does Ember 1 cost?
- Ember 1 costs $3.00 per million input tokens and $15.00 per million output tokens. Adjusted for how many tokens it actually consumes on real tasks, its effective cost works out to $15.87 per million input-plus-output tokens.
- Is Ember 1 open source?
- No — Ember 1 is a proprietary model. Its weights are not publicly available, so it is used through hosted APIs and products.
- Who developed Ember 1?
- Ember 1 was developed by Fireworks, based in the United States.
- When was Ember 1 released?
- Ember 1 was publicly released on September 23, 2026.
- How large is the context window of Ember 1?
- Ember 1 supports a context window of 1,048,576 tokens — the amount of text it can consider in a single request.
- How fast is Ember 1?
- AI IQ doesn't have independent speed measurements for Ember 1 yet. Speed comes from independent tests of a model's hosted API, which aren't available for every model — especially new, restricted-access, or self-hosted ones.