Celeris-1 vs Ling 3.0 Tiny
Wins 0 of 5 areas
—
Wins 4 of 5 areas
Coding · Agents · Reasoning · Long documents
Ling 3.0 Tiny is the stronger all-rounder.
Scores updated · 14 tests both models report · How we compare
Where each one wins
Tests won in each of the five areas where both have results. Each piece is one test, so a longer bar means more evidence; grey means the two scored within a point of each other.
- AgentsCarrying out multi-step tasks on its own02Ling 3.0 Tiny2 of 3 tests · 1 tie
- ReasoningHard problems that need careful thinking02Ling 3.0 Tiny2 of 3 tests · 1 tie
- CodingWriting and fixing software02Ling 3.0 Tiny2 of 2 tests
- Long documentsFinding answers in very long texts01Ling 3.0 Tiny1 of 1 test
- FactsGetting facts right instead of making them up11Even1 each
Long documents rests on a single test.
What it costs
Prices per million tokens, roughly 750,000 words. The bars show the cost of a million tokens read plus a million written.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where Ling 3.0 Tiny pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+62.3points ahead
- Reasons across sets of long documentsAA-LCR+22.3points ahead
- Customer-service tasks in a simulated bankτ-Bench V3 · Banking+16.9points ahead
Where Celeris-1 pulls ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+2.5points ahead
Every test, side by side
All 14 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingLing 3.0 Tiny
- Terminal-Bench 2.1Ling 3.0 Tiny by 16.511.227.7+16.5
- SciCodeLing 3.0 Tiny by 2.621.624.2+2.6
AgentsLing 3.0 Tiny
- τ-Bench V3 · BankingLing 3.0 Tiny by 16.93.920.8+16.9
- GDPValLing 3.0 Tiny by 3.203.2+3.2
- Terminal-Bench 4.0tie00tie
ReasoningLing 3.0 Tiny
- GPQA DiamondLing 3.0 Tiny by 10.363.173.4+10.3
- Humanity's Last ExamLing 3.0 Tiny by 2.56.89.3+2.5
- CritPttie00tie
FactsEven
- AA-Omniscience · Non-hallucinationLing 3.0 Tiny by 62.37.269.5+62.3
- AA-Omniscience · AccuracyCeleris-1 by 2.5118.5+2.5
Other results3 tests, not counted
Tests outside the eight areas. They are not counted above: several are summary scores built from other tests, or the same test under another name.
- AA-OmniscienceLing 3.0 Tiny by 52.3-71.6-19.4+52.3
- Artificial Analysis Coding IndexLing 3.0 Tiny by 12.114.426.5+12.1
- AA IntelligenceLing 3.0 Tiny by 4.86.311.1+4.8
Questions people ask
Which is better, Celeris-1 or Ling 3.0 Tiny?
Ling 3.0 Tiny wins four of the five areas where both have results: coding, agents, reasoning and long documents. Celeris-1 wins none. They are level on facts.
Which is better for coding?
Ling 3.0 Tiny. It wins 2 of the 2 coding tests both models report; Celeris-1 wins none.
How do you compare the two?
We use the 14 benchmark tests both models have published scores on. The verdict counts the 11 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 3 are listed but not counted, because several are summary scores or repeat a test. Each score is the one shown on the model's own page.