Independent model analysis
Which model, and what does it cost you?
404 models from 59 creators, measured on intelligence, output speed, latency and the price of getting real work done. No vendor benchmarks, no marketing numbers.
Top Intelligence Index
58.1
Qwen3.8 Max · Qwen
Fastest output
6,866 t/s
Apply 3
Cheapest benchmarked
$0.015
Ling-2.6-flash · per 1M tokens
Models benchmarked
198
of 404 indexed
Highlights
Intelligence, speed and cost
The three numbers that decide most model choices. Every chart on this site ships a table view; hover a bar for the rest of a model's profile.
Intelligence
Intelligence Index · higher is better
- Qwen3.8 MaxQwen58.1
- Claude Opus 5.5Anthropic57.6
- Claude Sonnet 5.5Anthropic56.0
- Claude Fable 5.1Anthropic53.4
- GPT-6 AstraOpenAI52.7
- Gemini 4 ArgonGoogle52.6
- Muse WonderMeta52.4
- GPT-6.1 SolOpenAI51.8
- Claude Opus 5Anthropic50.8
- Claude Fable 5Anthropic49.6
Composite of ten reasoning, knowledge, coding and agentic evaluations.
View as table
| Model | Index |
|---|---|
| Qwen3.8 Max (Qwen) | 58.1 |
| Claude Opus 5.5 (Anthropic) | 57.6 |
| Claude Sonnet 5.5 (Anthropic) | 56.0 |
| Claude Fable 5.1 (Anthropic) | 53.4 |
| GPT-6 Astra (OpenAI) | 52.7 |
| Gemini 4 Argon (Google) | 52.6 |
| Muse Wonder (Meta) | 52.4 |
| GPT-6.1 Sol (OpenAI) | 51.8 |
| Claude Opus 5 (Anthropic) | 50.8 |
| Claude Fable 5 (Anthropic) | 49.6 |
Speed
Output tokens per second · higher is better
- Apply 3Relace6,866 t/s
- V3 LargeMorph3,208 t/s
- V3 FastMorph3,005 t/s
- gpt-oss-120bOpenAI641 t/s
- gpt-oss-safeguard-20bOpenAI537 t/s
- 461 t/s
- gpt-oss-20bOpenAI410 t/s
- Qwen3.8 27BQwen409 t/s
- 406 t/s
- GLM 5.2Z.ai368 t/s
Median output tokens per second across serving providers.
View as table
| Model | Tokens/s |
|---|---|
| Apply 3 (Relace) | 6866 |
| V3 Large (Morph) | 3208 |
| V3 Fast (Morph) | 3005 |
| gpt-oss-120b (OpenAI) | 641 |
| gpt-oss-safeguard-20b (OpenAI) | 537 |
| Llama 3.1 8B Instruct (Meta) | 461 |
| gpt-oss-20b (OpenAI) | 410 |
| Qwen3.8 27B (Qwen) | 409 |
| Grok 4.20 Multi-Agent (xAI) | 406 |
| GLM 5.2 (Z.ai) | 368 |
Cost per task
Estimated USD per task · lower is better
- V4 Flash 0423DeepSeek$0.02
- Ling 3.1 FlashInclusionAI$0.02
- MiMo-V2.6-FlashXiaomi$0.03
- Claude Haiku 5.5Anthropic$0.04
- GPT-6 LunaOpenAI$0.04
- KAT-Coder-Pro V2KwaiPilot$0.04
- GLM 5.3 FlashZ.ai$0.05
- V4 Flash 0731DeepSeek$0.05
- Ring-2.6-1TInclusionAI$0.05
- V4.1 FlashDeepSeek$0.07
Estimated from list pricing: 50K input tokens plus 80K output tokens for reasoning models (25K for non-reasoning). Limited to models scoring 30+ on the Intelligence Index, so the ranking compares like with like.
View as table
| Model | USD |
|---|---|
| V4 Flash 0423 (DeepSeek) | $0.02 |
| Ling 3.1 Flash (InclusionAI) | $0.02 |
| MiMo-V2.6-Flash (Xiaomi) | $0.03 |
| Claude Haiku 5.5 (Anthropic) | $0.04 |
| GPT-6 Luna (OpenAI) | $0.04 |
| KAT-Coder-Pro V2 (KwaiPilot) | $0.04 |
| GLM 5.3 Flash (Z.ai) | $0.05 |
| V4 Flash 0731 (DeepSeek) | $0.05 |
| Ring-2.6-1T (InclusionAI) | $0.05 |
| V4.1 Flash (DeepSeek) | $0.07 |
Latency
Seconds to first token · lower is better
- 64ms
- Gemma 4 31BGoogle89ms
- Phi 4Microsoft98ms
- Granite 4.1 8BIBM Granite118ms
- Llama 3 8B LunarisSao10K143ms
- 151ms
- 168ms
- gpt-oss-safeguard-20bOpenAI177ms
- Nemotron 3.5 LightningNVIDIA194ms
- Mistral Small 4Mistral AI194ms
Median time to first token.
View as table
| Model | Seconds |
|---|---|
| Llama 3.1 8B Instruct (Meta) | 64ms |
| Gemma 4 31B (Google) | 89ms |
| Phi 4 (Microsoft) | 98ms |
| Granite 4.1 8B (IBM Granite) | 118ms |
| Llama 3 8B Lunaris (Sao10K) | 143ms |
| Llama 3.3 70B Instruct (Meta) | 151ms |
| Nemotron 3.5 Content Safety (NVIDIA) | 168ms |
| gpt-oss-safeguard-20b (OpenAI) | 177ms |
| Nemotron 3.5 Lightning (NVIDIA) | 194ms |
| Mistral Small 4 (Mistral AI) | 194ms |
Blended price
USD per 1M tokens, 3:1 input:output · lower is better
- Ling-2.6-flashInclusionAI$0.015
- Ling 3.0 FlashInclusionAI$0.032
- gpt-oss-20bOpenAI$0.036
- Granite 4.1 8BIBM Granite$0.063
- Mercury 2.5Inception$0.068
- Solar Mini 4Upstage$0.088
- Phi 4Microsoft$0.088
- Ling 3.0 Flash VLInclusionAI$0.09
- Nemotron 3.5 LightningNVIDIA$0.103
- Nemotron 3 Nano 30B A3BNVIDIA$0.105
3:1 weighted blend of input and output list price. Free endpoints are excluded; only benchmarked models appear.
View as table
| Model | USD / 1M |
|---|---|
| Ling-2.6-flash (InclusionAI) | $0.015 |
| Ling 3.0 Flash (InclusionAI) | $0.032 |
| gpt-oss-20b (OpenAI) | $0.036 |
| Granite 4.1 8B (IBM Granite) | $0.063 |
| Mercury 2.5 (Inception) | $0.068 |
| Solar Mini 4 (Upstage) | $0.088 |
| Phi 4 (Microsoft) | $0.088 |
| Ling 3.0 Flash VL (InclusionAI) | $0.09 |
| Nemotron 3.5 Lightning (NVIDIA) | $0.103 |
| Nemotron 3 Nano 30B A3B (NVIDIA) | $0.105 |
Trade-offs
Nothing is free
Intelligence, speed and price pull against each other. These two plots are where the actual decision gets made.
Intelligence vs price
Blended USD per 1M tokens on a log scale · up and to the right is a better deal
- Proprietary
- Open weights
Price axis is inverted so cheaper sits right. Labelled points are the Pareto frontier — nothing in the index is both smarter and cheaper. Only models with both a published price and an Intelligence Index score appear.
View as table
| Model | Intelligence | Blended / 1M |
|---|---|---|
| Qwen3.8 Max (Qwen) | 58.1 | $3 |
| Claude Opus 5.5 (Anthropic) | 57.6 | $8 |
| Claude Sonnet 5.5 (Anthropic) | 56.0 | $4 |
| Claude Fable 5.1 (Anthropic) | 53.4 | $20 |
| GPT-6 Astra (OpenAI) | 52.7 | $20 |
| Gemini 4 Argon (Google) | 52.6 | $4 |
| Muse Wonder (Meta) | 52.4 | $8 |
| GPT-6.1 Sol (OpenAI) | 51.8 | $4 |
| Claude Opus 5 (Anthropic) | 50.8 | $10 |
| Claude Fable 5 (Anthropic) | 49.6 | $20 |
| Muse Spark 1.3 (Meta) | 48.1 | $2 |
| GPT-6 Sol (OpenAI) | 47.6 | $4 |
| GPT-5.6 Sol (OpenAI) | 47.0 | $8 |
| Grok 4.7 (xAI) | 46.4 | $3 |
| MiMo-V2.6-Pro (Xiaomi) | 46.3 | $0.544 |
| Qwen3.8 Max (0902) (Qwen) | 45.4 | $3 |
| GLM 5.3 (Z.ai) | 44.8 | $1.22 |
| Grok 4.6 (xAI) | 44.3 | $3 |
| Step 5 Preview (StepFun) | 43.7 | $1.43 |
| Kimi K3 (Moonshot AI) | 43.6 | $4.09 |
| Claude Haiku 5.5 (Anthropic) | 43.4 | $0.2 |
| GPT-5.6 Terra (OpenAI) | 42.1 | $4.5 |
| GLM 5.3 Flash (Z.ai) | 41.8 | $0.237 |
| Claude Opus 4.8 (Anthropic) | 41.8 | $10 |
| Nex-N2-Pro (Nex AGI) | 41.7 | $0.438 |
Intelligence vs output speed
Median tokens per second · up and to the right is both smart and fast
- Proprietary
- Open weights
Labelled points are the Pareto frontier — nothing is both smarter and faster. Output speed is the median across serving providers, so a model served by many reports a blended figure.
View as table
| Model | Intelligence | Tokens/s |
|---|---|---|
| Qwen3.8 Max (Qwen) | 58.1 | 38 |
| Claude Opus 5.5 (Anthropic) | 57.6 | 174 |
| Claude Sonnet 5.5 (Anthropic) | 56.0 | 118 |
| Claude Fable 5.1 (Anthropic) | 53.4 | 70 |
| GPT-6 Astra (OpenAI) | 52.7 | 150 |
| Gemini 4 Argon (Google) | 52.6 | 76 |
| Muse Wonder (Meta) | 52.4 | 74 |
| GPT-6.1 Sol (OpenAI) | 51.8 | 133 |
| Claude Opus 5 (Anthropic) | 50.8 | 98 |
| Claude Fable 5 (Anthropic) | 49.6 | 63 |
| Muse Spark 1.3 (Meta) | 48.1 | 65 |
| GPT-6 Sol (OpenAI) | 47.6 | 77 |
| GPT-5.6 Sol (OpenAI) | 47.0 | 121 |
| Grok 4.7 (xAI) | 46.4 | 71 |
| MiMo-V2.6-Pro (Xiaomi) | 46.3 | 39 |
| Qwen3.8 Max (0902) (Qwen) | 45.4 | 33 |
| GLM 5.3 (Z.ai) | 44.8 | 138 |
| Grok 4.6 (xAI) | 44.3 | 102 |
| Step 5 Preview (StepFun) | 43.7 | 49 |
| Kimi K3 (Moonshot AI) | 43.6 | 114 |
| Claude Haiku 5.5 (Anthropic) | 43.4 | 182 |
| GPT-5.6 Terra (OpenAI) | 42.1 | 102 |
| GLM 5.3 Flash (Z.ai) | 41.8 | 141 |
| Claude Opus 4.8 (Anthropic) | 41.8 | 181 |
| Nex-N2-Pro (Nex AGI) | 41.7 | 86 |
AI trends
The frontier, month by month
Best Intelligence Index score available at any point in time. The line steps only when a release actually beats the record.
Frontier Intelligence Index
Best score available on any given date · one series, so no legend is needed
A step means a new record holder; a flat run means nobody beat it.
View as table
| Date | Model | Index |
|---|---|---|
| May 2023 | GPT-4 | 6.7 |
| Apr 2024 | GPT-4 Turbo | 7.0 |
| May 2024 | GPT-4o | 8.4 |
| Dec 2024 | o1 | 15.2 |
| Apr 2025 | o4 Mini | 16.7 |
| Apr 2025 | o3 | 20.2 |
| May 2025 | Claude Opus 4 | 26.0 |
| Dec 2025 | GPT-5.2 | 30.4 |
| Feb 2026 | Qwen3 Max Thinking | 32.5 |
| Mar 2026 | GPT-5.4 | 39.0 |
| Apr 2026 | Claude Opus 4.7 | 40.7 |
| Apr 2026 | Qwen3.6 Max Preview | 41.1 |
| May 2026 | Claude Opus 4.8 | 41.8 |
| Jun 2026 | Claude Fable 5 | 49.6 |
| Jul 2026 | Claude Opus 5 | 50.8 |
| Aug 2026 | Qwen3.8 Max | 58.1 |
Explore
Seven ways into the data
Models
Every model in the index, filterable by capability, creator and price, with the full metric table.
OpenCoding Agents
Agentic and coding indices, terminal work, tool use, and head-to-head full-stack build quality.
OpenSpeech / Image / Video
Which models hear audio, read video, and emit images — and what that capability costs.
OpenInference
Serving providers ranked on output speed, time to first token and blended price.
OpenLeaderboards
One ranked board per benchmark: GPQA, Humanity's Last Exam, SciCode, τ²-bench, long context.
OpenAI Trends
How the frontier moved, and how fast the price of a given intelligence level fell.
OpenArena
Pairwise Elo across websites, UI components, data visualisation, 3D and game development.
OpenStart from the number that matters to you
Cheapest model that clears a quality bar, fastest model that can still use tools, or the single best score on one benchmark — every ranking is one click away.
Cost per task figures are estimates derived from list pricing — see methodology. Example: $0.58 for Qwen3.8 Max.