AI model
performance
report
A concise view of coding completion, cost efficiency, model popularity, and inference provider reliability across the Kilo ecosystem.
- Published
- October 15, 2026
- Benchmark
- Terminal Bench 2.0
- Inference window
- Latest weekly snapshot
KiloBench leaders
Models are ranked by completion rate on Terminal Bench 2.0. Cost per attempt is the observed average for a complete benchmark attempt, including unsuccessful attempts.
| # | Model | Provider | Completion rate | Cost / attempt |
|---|---|---|---|---|
| 1 | GPT-6 Astra ($$$$) 5 attempts | openai | 79.3% | $107.29 |
| 2 | GPT-5.6 Sol 5 attempts | openai | 76.2% | $87.41 |
| 3 | Claude Fable 5.1 ($$$$) 5 attempts | anthropic | 76.2% | $91.41 |
| 4 | DeepSeek V4.1 Flash 5 attempts | deepseek | 75.3% | $2.58 |
| 5 | Gemini 3.8 Flash (50% off) 5 attempts | 75.3% | $112.27 | |
| 6 | GPT-5.5 5 attempts | openai | 74.2% | $72.63 |
| 7 | Grok 4.6 5 attempts | x-ai | 73.0% | $33.83 |
| 8 | Kimi K3 5 attempts | moonshotai | 72.8% | $48.38 |
| 9 | Claude Opus 5 5 attempts | anthropic | 71.5% | $113.54 |
| 10 | Claude Fable 5 ($$$$) 5 attempts | anthropic | 71.0% | $87.52 |
Inference market view
The 20 most-used models are ranked by token volume from the latest Kilo Gateway weekly snapshot. Primary-provider metrics describe the provider carrying the largest traffic share for each model.
| # | Model | Primary provider | Usage volume | Cost / request | Success rate |
|---|---|---|---|---|---|
| 1 | step-5-preview-free 1 provider | StepFun | 2.0T | $0 | 98.59% |
| 2 | nemotron-3-ultra-550b-a55b:free 1 provider | NVIDIA | 1.4T | $0 | 93.82% |
| 3 | dots-3-note-preview:free 1 provider | Atlas Cloud | 598.8B | $0 | 85.93% |
| 4 | laguna-s-2.1:free 1 provider | Poolside | 566.5B | $0 | 95.19% |
| 5 | nemotron-3-super-120b-a12b:free 1 provider | NVIDIA | 275.0B | $0 | 91.66% |
| 6 | deepseek-v4.1-flash 6 providers | Novita AI | 220.5B | $0.0029 | 99.63% |
| 7 | glyph-cluster 1 provider | Stealth | 199.2B | $0 | 97.04% |
| 8 | ling-3.1-flash 1 provider | Novita AI | 158.0B | $0.00001 | 90.81% |
| 9 | openrouter/free 8 providers | NVIDIA | 93.5B | $0 | 86.56% |
| 10 | nemotron-3.5-lightning:free 1 provider | NVIDIA | 82.5B | $0 | 77.36% |
Inference market view / continued
The 20 most-used models are ranked by token volume from the latest Kilo Gateway weekly snapshot. Primary-provider metrics describe the provider carrying the largest traffic share for each model.
| # | Model | Primary provider | Usage volume | Cost / request | Success rate |
|---|---|---|---|---|---|
| 11 | north-mini-code:free 1 provider | Cohere | 53.9B | $0 | 82.21% |
| 12 | glm-5.3-flash 6 providers | Friendli AI | 51.0B | $0.0062 | 99.83% |
| 13 | step-3.7-flash:free 1 provider | StepFun | 47.3B | $0 | 96.23% |
| 14 | gpt-6-luna 3 providers | OpenAI | 43.6B | $0.0029 | 99.94% |
| 15 | mimo-v2.6-pro 3 providers | Xiaomi | 38.8B | $0.0037 | 99.62% |
| 16 | claude-opus-5.5 3 providers | 38.4B | $0.12 | 95.02% | |
| 17 | muse-spark-1.3-contributor 1 provider | Meta | 35.6B | $0.0016 | 99.81% |
| 18 | gpt-6.1-sol 3 providers | OpenAI | 35.6B | $0.0520 | 99.95% |
| 19 | ling-3.0-flash-sante:free 1 provider | Novita AI | 16.2B | $0 | 91.41% |
| 20 | claude-sonnet-5.5 3 providers | 15.9B | $0.0760 | 98.67% |
Cost and reliability
Provider-level traffic share, average request cost, and success rate for the ten most-used models. Traffic share uses the percentage published by Kilo Gateway.
step-5-preview-free
| Provider | Traffic share | Cost / request | Success rate |
|---|---|---|---|
| StepFun | 100.00% | $0 | 98.59% |
nemotron-3-ultra-550b-a55b:free
| Provider | Traffic share | Cost / request | Success rate |
|---|---|---|---|
| NVIDIA | 100.00% | $0 | 93.82% |
dots-3-note-preview:free
| Provider | Traffic share | Cost / request | Success rate |
|---|---|---|---|
| Atlas Cloud | 100.00% | $0 | 85.93% |
laguna-s-2.1:free
| Provider | Traffic share | Cost / request | Success rate |
|---|---|---|---|
| Poolside | 100.00% | $0 | 95.19% |
nemotron-3-super-120b-a12b:free
| Provider | Traffic share | Cost / request | Success rate |
|---|---|---|---|
| NVIDIA | 100.00% | $0 | 91.66% |
Cost and reliability
Provider-level traffic share, average request cost, and success rate for the ten most-used models. Traffic share uses the percentage published by Kilo Gateway.
deepseek-v4.1-flash
| Provider | Traffic share | Cost / request | Success rate |
|---|---|---|---|
| Novita AI | 94.07% | $0.0029 | 99.63% |
| Alibaba | 4.50% | $0.0146 | 99.87% |
| Fireworks AI | 0.98% | $0.0046 | 99.93% |
| Together AI | 0.27% | $0.0035 | 100.00% |
| Baseten | 0.18% | $0.0086 | 99.92% |
| Relace | <0.01% | $0.0013 | 100.00% |
glyph-cluster
| Provider | Traffic share | Cost / request | Success rate |
|---|---|---|---|
| Stealth | 100.00% | $0 | 97.04% |
ling-3.1-flash
| Provider | Traffic share | Cost / request | Success rate |
|---|---|---|---|
| Novita AI | 100.00% | $0.00001 | 90.81% |
Cost and reliability
Provider-level traffic share, average request cost, and success rate for the ten most-used models. Traffic share uses the percentage published by Kilo Gateway.
openrouter/free
| Provider | Traffic share | Cost / request | Success rate |
|---|---|---|---|
| NVIDIA | 48.05% | $0 | 86.56% |
| Atlas Cloud | 15.49% | $0 | 82.58% |
| Novita AI | 13.83% | $0 | 81.70% |
| Poolside | 13.60% | $0 | 93.24% |
| Cohere | 7.57% | $0 | 80.40% |
| Liquid | 1.19% | $0 | 58.51% |
| Thinking Machines | 0.26% | $0 | 86.38% |
| 0.01% | $0 | 91.03% |
nemotron-3.5-lightning:free
| Provider | Traffic share | Cost / request | Success rate |
|---|---|---|---|
| NVIDIA | 100.00% | $0 | 77.36% |