Ember-1 vs Nemotron 3.5 Super VL (120B A12B): Specs & Benchmark Comparison

Ember-1 is developed by Fireworks AI, while Nemotron 3.5 Super VL (120B A12B) comes from NVIDIA. Ember-1 was released in September 2026, and Nemotron 3.5 Super VL (120B A12B) followed a month later in October 2026. Ember-1 is the larger model at roughly 2.8 trillion parameters, against 120 billion for Nemotron 3.5 Super VL (120B A12B).

The two models share 2 published benchmarks. Ember-1 leads on 2 of them. The widest gaps are on SWE-Bench Verified, where Ember-1 scores 92.2% against 71.1%; Terminal-Bench 2.1, where Ember-1 scores 82.0% against 66.7%. Averaged across everything we track, Ember-1 sits at 78.8% and Nemotron 3.5 Super VL (120B A12B) at 60.9%.

Nemotron 3.5 Super VL (120B A12B) is the cheaper API at $0.20 per million input tokens and $1 per million output tokens, roughly 15 times cheaper than Ember-1 at $3 and $15. Ember-1 takes the larger context window at 1M tokens, compared with 262K for Nemotron 3.5 Super VL (120B A12B). Ember-1 accepts text as input, while Nemotron 3.5 Super VL (120B A12B) accepts text, images, and video.

CharacteristicEmber-1Nemotron 3.5 Super VL (120B A12B)
CompanyFireworks AINVIDIA
Release DateSeptember 23, 2026October 15, 2026
Parameters2.8T120B
MultimodalNoYes
Context (input)1.0M262K
Context (output)—262K
Input Price / 1M$3.00$0.20
Output Price / 1M$15.00$1.00
Average Score78.8%60.9%
Benchmarks
SWE-Bench Verified92.2%71.1%
Terminal-Bench 2.182.0%66.7%

Visual Benchmark Comparison

Ember-1
Nemotron 3.5 Super VL (120B A12B)
SWE-Bench Verified0.9 vs 0.7
0.9
0.7
Terminal-Bench 2.10.8 vs 0.7
0.8
0.7

Verdict

Nemotron 3.5 Super VL (120B A12B) leads in 2 out of 5 comparison categories.

Overall Performance

Both models show comparable average scores: Ember-1 — 0.8, Nemotron 3.5 Super VL (120B A12B) — 0.6.

Programming

On SWE-Bench, both models are nearly equal: Ember-1 — 0.9, Nemotron 3.5 Super VL (120B A12B) — 0.7.

API Cost

Nemotron 3.5 Super VL (120B A12B) is 15.0x cheaper: input $0.20/1M vs $3.00/1M tokens.

Context Window

Ember-1 supports a larger context: 1M vs 262K tokens.

Recency

Nemotron 3.5 Super VL (120B A12B) is newer: released 10/15/2026 vs 9/23/2026.

More About These Models

Related Comparisons

Frequently Asked Questions

Which is better for coding — Ember-1 or Nemotron 3.5 Super VL (120B A12B)?
On the SWE-Bench benchmark, Ember-1 shows a better result: 92.2% vs 71.1%.
Which model is cheaper — Ember-1 or Nemotron 3.5 Super VL (120B A12B)?
Nemotron 3.5 Super VL (120B A12B) is cheaper for input: $0.20 per 1M tokens vs $3.00.
Which has a larger context window — Ember-1 or Nemotron 3.5 Super VL (120B A12B)?
Ember-1 supports a larger context: 1,040,000 tokens vs 262,144.

The Ember-1 and Nemotron 3.5 Super VL (120B A12B) comparison is updated for 2026. Data includes benchmark results, API pricing, context window size and other specifications. For more detailed information, visit the Ember-1 or Nemotron 3.5 Super VL (120B A12B) page. See also the complete list of AI model comparisons.