# LLM Leaderboard & AI Model Benchmarks — September 2026

> Compare frontier AI models by quality, cost, and context. 232 ranked and 130 verified models among 417 tracked LLMs across 422 benchmark records and 8 weighted categories.

**Last verified:** September 4, 2026

Canonical page: https://benchlm.ai/

## Current answer

> As of September 4, 2026, Claude Fable 5.1 leads the BenchAlign BenchLM leaderboard with a score of 82.95. Gemini 3.8 Flash retains 95% of that score at $3.75 per 1M output tokens, 93% below the leader's $50 price.

## Ranking coverage

- **Provisional-ranked lane:** 232 ranked entries in the current snapshot. This lane can include public, non-generated rows while exact citations are still being attached.
- **Verified-ranked coverage:** 130 models meet the exact-source ranking requirements.
- **Method:** BenchLM never interpolates missing scores. See the [scoring methodology](/methodology) and [evidence confidence guide](/benchmark-confidence).

## Decision-ready picks

| Decision | Model | Current measure | Provenance |
|----------|-------|-----------------|------------|
| Best open weight | [Qwen3.8 Max](/models/qwen3-8-max) | 72.43 overall score | Supported · 3 source families |
| Best near-frontier value | [Gemini 3.8 Flash](/models/gemini-3-8-flash) | $3.75 output / 1M tokens | Supported · 2 source families |
| Fastest measured | [LFM2.5-2.6B](/models/lfm2-5-2-6b) | 595 tokens / sec | Artificial Analysis · updated 2026-09-04 |
| Largest useful context | [Grok 4.20](/models/grok-4-20-beta) | 2M context window | Supported · 2 source families |

## Overall rankings

| Rank | Model | Provider | Type | Context | Score | Output price / 1M | Evidence |
|------|-------|----------|------|---------|-------|-------------------|----------|
| 1 | [Claude Fable 5.1](/models/claude-fable-5-1) | Anthropic | Proprietary | 1M | 82.95 | $50.00 | Supported-ranked (3 source families) |
| 2 | [GPT-6 Astra](/models/gpt-6-astra) | OpenAI | Proprietary | 1.05M | 81.05 | $50.00 | Estimated-ranked (2 source families) |
| 3 | [Claude Fable 5](/models/claude-fable) | Anthropic | Proprietary | 1M+ | 80.91 | $50.00 | Supported-ranked (3 source families) |
| 4 | [Claude Opus 5](/models/claude-opus-5) | Anthropic | Proprietary | Not listed | 80.67 | $25.00 | Supported-ranked (3 source families) |
| 5 | [GPT-5.6 Sol](/models/gpt-5-6-sol) | OpenAI | Proprietary | 1.05M | 79.65 | $30.00 | Supported-ranked (3 source families) |
| 6 | [Gemini 3.8 Flash](/models/gemini-3-8-flash) | Google | Proprietary | 1M | 78.41 | $3.75 | Supported-ranked (2 source families) |
| 7 | [Kimi K3](/models/kimi-k3) | Moonshot AI | Pending | 1.05M | 74.87 | $15.00 | Supported-ranked (3 source families) |
| 8 | [GPT-5.5](/models/gpt-5-5) | OpenAI | Proprietary | 1M | 73.27 | $30.00 | Supported-ranked (3 source families) |
| 9 | [Qwen3.8 Max](/models/qwen3-8-max) | Alibaba | Open Weight | 1M | 72.43 | Not listed | Supported-ranked (3 source families) |
| 10 | [Claude Opus 4.8](/models/claude-opus-4-8) | Anthropic | Proprietary | 1M | 72.3 | $25.00 | Supported-ranked (3 source families) |
| 11 | [Muse Spark 1.2](/models/muse-spark-1-2) | Meta | Proprietary | 1M | 71.88 | $4.25 | Estimated-ranked (2 source families) |
| 12 | [Muse Spark 1.1](/models/muse-spark-1-1) | Meta | Proprietary | 1M | 71.78 | Not listed | Supported-ranked (3 source families) |
| 13 | [GPT-5.6 Terra](/models/gpt-5-6-terra) | OpenAI | Proprietary | 1.05M | 71.44 | $15.00 | Estimated-ranked (2 source families) |
| 14 | [GPT-5.4](/models/gpt-5-4) | OpenAI | Proprietary | 1.05M | 70.96 | $15.00 | Supported-ranked (3 source families) |
| 15 | [Claude Sonnet 5](/models/claude-sonnet-5) | Anthropic | Proprietary | 1M | 70.76 | $10.00 | Supported-ranked (3 source families) |
| 16 | [Claude Opus 4.7](/models/claude-opus-4-7) | Anthropic | Proprietary | 1M | 70.22 | $25.00 | Supported-ranked (2 source families) |
| 17 | [Gemini 3.7 Flash](/models/gemini-3-7-flash) | Google | Proprietary | 1M | 70.21 | $3.75 | Estimated-ranked (2 source families) |
| 18 | [Grok 4.6](/models/grok-4-6) | xAI | Proprietary | 500K | 70.19 | $6.00 | Supported-ranked (3 source families) |
| 19 | [Gemini 3.6 Flash](/models/gemini-3-6-flash) | Google | Proprietary | 1M | 70.11 | $7.50 | Supported-ranked (3 source families) |
| 20 | [Gemini 3.5 Flash](/models/gemini-3-5-flash) | Google | Proprietary | 1M | 70.01 | $9.00 | Supported-ranked (3 source families) |
| 21 | [Claude Opus 4.7 (Adaptive)](/models/claude-opus-4-7-adaptive) | Anthropic | Proprietary | 1M | 69.94 | $25.00 | Estimated-ranked (1 source family) |
| 22 | [Claude Opus 4.6](/models/claude-opus-4-6) | Anthropic | Proprietary | 1M | 69.84 | $25.00 | Supported-ranked (3 source families) |
| 23 | [Gemini 3.1 Pro](/models/gemini-3-1-pro) | Google | Proprietary | 1M | 69.82 | $12.00 | Supported-ranked (2 source families) |
| 24 | [GPT-5.2 Pro](/models/gpt-5-2-pro) | OpenAI | Proprietary | 400K | 68.73 | $168.00 | Supported-ranked (2 source families) |
| 25 | [Grok 4.5](/models/grok-4-5) | xAI | Proprietary | 500K | 68.64 | $6.00 | Supported-ranked (3 source families) |
| 26 | [Qwen3.7 Max](/models/qwen3-7-max) | Alibaba | Proprietary | 1M | 68.56 | Not listed | Supported-ranked (3 source families) |
| 27 | [GLM-5.3](/models/glm-5-3) | Z.AI | Open Weight | 1M | 68.47 | Self-hosted / varies | Estimated-ranked (2 source families) |
| 28 | [Muse Spark](/models/muse-spark) | Meta | Proprietary | 262K | 68.39 | Not listed | Supported-ranked (3 source families) |
| 29 | [Qwen3.8-27B](/models/qwen3-8-27b) | Alibaba | Open Weight | 262K | 68.35 | Self-hosted / varies | Supported-ranked (2 source families) |
| 30 | [GLM-5.2](/models/glm-5-2) | Z.AI | Open Weight | 1M | 68.22 | $4.40 | Supported-ranked (3 source families) |
| 31 | [Gemini 3 Pro](/models/gemini-3-pro) | Google | Proprietary | 2M | 67.24 | $12.00 | Supported-ranked (3 source families) |
| 32 | [Grok 4.20](/models/grok-4-20-beta) | xAI | Proprietary | 2M | 67.01 | $6.00 | Supported-ranked (2 source families) |
| 33 | [DeepSeek V4 Pro 0813](/models/deepseek-v4-pro-0813) | DeepSeek | Proprietary | 1M | 66.35 | $0.87 | Estimated-ranked (1 source family) |
| 34 | [GLM-5.3-Flash](/models/glm-5-3-flash) | Z.AI | Open Weight | 1M | 66.09 | Self-hosted / varies | Supported-ranked (2 source families) |
| 35 | [Hy4 preview](/models/hy4-preview) | Tencent | Open Weight | 1M | 65.94 | Self-hosted / varies | Estimated-ranked (9 source families) |
| 36 | [Hy3](/models/hy3) | Tencent | Open Weight | 256K | 65.84 | Self-hosted / varies | Supported-ranked (2 source families) |
| 37 | [GPT-5.6 Luna](/models/gpt-5-6-luna) | OpenAI | Proprietary | 1.05M | 65.54 | $6.00 | Estimated-ranked (2 source families) |
| 38 | [GPT-5.2](/models/gpt-5-2) | OpenAI | Proprietary | 400K | 65.45 | $14.00 | Supported-ranked (3 source families) |
| 39 | [MiMo-V2.5-Pro](/models/mimo-v2-5-pro) | Xiaomi | Proprietary | 1M | 65.41 | Not listed | Supported-ranked (3 source families) |
| 40 | [Kimi K2.7 Code](/models/kimi-k2-7-code) | Moonshot AI | Open Weight | 256K | 65.4 | $4.00 | Estimated-ranked (2 source families) |
| 41 | [Kimi K2.6](/models/kimi-2-6) | Moonshot AI | Open Weight | 256K | 65.3 | $4.00 | Supported-ranked (2 source families) |
| 42 | [Ornith-1.5-397B](/models/ornith-1-5-397b) | Ornith AI | Open Weight | 262K | 65.25 | Self-hosted / varies | Estimated-ranked (8 source families) |
| 43 | [GPT-5.3 Codex](/models/gpt-5-3-codex) | OpenAI | Proprietary | 400K | 65.22 | $14.00 | Supported-ranked (3 source families) |
| 44 | [MiMo-V2-Pro](/models/mimo-v2-pro) | Xiaomi | Proprietary | 1M | 64.9 | Not listed | Supported-ranked (2 source families) |
| 45 | [GLM-5.1](/models/glm-5-1) | Z.AI | Open Weight | 203K | 64.41 | $4.40 | Supported-ranked (3 source families) |
| 46 | [Claude Sonnet 4.6](/models/claude-sonnet-4-6) | Anthropic | Proprietary | 200K | 64.21 | $15.00 | Supported-ranked (3 source families) |
| 47 | [GLM-5-Turbo](/models/glm-5-turbo) | Z.AI | Proprietary | 200K | 64.14 | $4.00 | Supported-ranked (2 source families) |
| 48 | [dots3-note Preview](/models/dots3-note-preview) | Dots Studio | Open Weight | 512K | 64.09 | Self-hosted / varies | Estimated-ranked (9 source families) |
| 49 | [MiniMax M3](/models/minimax-m3) | MiniMax | Open Weight | 1M | 63.91 | $1.20 | Supported-ranked (3 source families) |
| 50 | [GPT-5.1](/models/gpt-5-1) | OpenAI | Proprietary | 200K | 63.69 | $10.00 | Supported-ranked (3 source families) |
| 51 | [Qwen 3.6 Max (preview)](/models/qwen3-6-max-preview) | Alibaba | Proprietary | 256K | 63.64 | Not listed | Supported-ranked (2 source families) |
| 52 | [Claude Opus 4.6 (Adaptive)](/models/claude-opus-4-6-thinking) | Anthropic | Proprietary | 1M | 63.58 | Not listed | Estimated-ranked (1 source family) |
| 53 | [GPT-5.5 Pro](/models/gpt-5-5-pro) | OpenAI | Proprietary | 1M | 63.08 | $180.00 | Estimated-ranked (1 source family) |
| 54 | [Gemini 3 Flash](/models/gemini-3-flash) | Google | Proprietary | 1M | 62.96 | $3.00 | Supported-ranked (3 source families) |
| 55 | [Inkling](/models/inkling) | Thinking Machines Lab | Open Weight | 1M | 62.78 | $4.68 | Supported-ranked (3 source families) |
| 56 | [GPT-5.4 mini](/models/gpt-5-4-mini) | OpenAI | Proprietary | 400K | 62.63 | $4.50 | Supported-ranked (3 source families) |
| 57 | [Qwen3.7 Plus](/models/qwen3-7-plus) | Alibaba | Proprietary | 1M | 62.3 | Not listed | Supported-ranked (3 source families) |
| 58 | [GPT-5.4 nano](/models/gpt-5-4-nano) | OpenAI | Proprietary | 400K | 62.19 | $1.25 | Supported-ranked (3 source families) |
| 59 | [GLM-5](/models/glm-5) | Z.AI | Open Weight | 200K | 62.16 | $3.20 | Supported-ranked (3 source families) |
| 60 | [GPT-5.4 Pro](/models/gpt-5-4-pro) | OpenAI | Proprietary | 1.05M | 61.76 | $180.00 | Estimated-ranked (1 source family) |
| 61 | [Qwen3.6 Plus](/models/qwen3-6-plus) | Alibaba | Proprietary | 1M | 61.49 | Not listed | Supported-ranked (3 source families) |
| 62 | [Inkling-Small](/models/inkling-small) | Thinking Machines Lab | Open Weight | 1M | 60.93 | $1.44 | Supported-ranked (3 source families) |
| 63 | [MiMo-V2.5](/models/mimo-v2-5) | Xiaomi | Proprietary | 1M | 60.72 | Not listed | Estimated-ranked (1 source family) |
| 64 | [Gemini 3 Pro Deep Think](/models/gemini-3-pro-deep-think) | Google | Proprietary | 2M | 60.67 | Not listed | Estimated-ranked (1 source family) |
| 65 | [MiMo-V2-Omni](/models/mimo-v2-omni) | Xiaomi | Proprietary | 262K | 60.62 | Not listed | Supported-ranked (2 source families) |
| 66 | [GLM-5V-Turbo](/models/glm-5v-turbo) | Z.AI | Proprietary | 200K | 60.56 | $4.00 | Supported-ranked (2 source families) |
| 67 | [Gemini 3.5 Flash-Lite](/models/gemini-3-5-flash-lite) | Google | Proprietary | 1M | 60.5 | $2.50 | Supported-ranked (3 source families) |
| 68 | [Grok 4.3](/models/grok-4-3) | xAI | Proprietary | 1M | 60.38 | $2.50 | Supported-ranked (3 source families) |
| 69 | [Claude Opus 4.5](/models/claude-opus-4-5) | Anthropic | Proprietary | 200K | 60.25 | $25.00 | Supported-ranked (3 source families) |
| 70 | [Qwen3.8-Flash-Next](/models/qwen3-8-flash-next) | Alibaba | Open Weight | 262K | 59.43 | Self-hosted / varies | Estimated-ranked (1 source family) |
| 71 | [Grok 4.1](/models/grok-4-1) | xAI | Proprietary | 1M | 59.34 | Not listed | Estimated-ranked (1 source family) |
| 72 | [GLM-5 (Reasoning)](/models/glm-5-reasoning) | Z.AI | Open Weight | 200K | 59.16 | $3.20 | Estimated-ranked (1 source family) |
| 73 | [Qwen3.5-122B-A10B](/models/qwen3-5-122b-a10b) | Alibaba | Open Weight | 262K | 59.01 | Self-hosted / varies | Supported-ranked (3 source families) |
| 74 | [GLM-4.7](/models/glm-4-7) | Z.AI | Open Weight | 200K | 58.99 | Self-hosted / varies | Supported-ranked (3 source families) |
| 75 | [MiniMax M2.7](/models/minimax-m2-7) | MiniMax | Open Weight | 200K | 58.99 | $1.20 | Supported-ranked (3 source families) |
| 76 | [Qwen3.5 397B (Reasoning)](/models/qwen3-5-397b-reasoning) | Alibaba | Open Weight | 128K | 58.89 | $3.60 | Estimated-ranked (1 source family) |
| 77 | [Kimi K2.5 (Reasoning)](/models/kimi-k2-5-reasoning) | Moonshot AI | Proprietary | 128K | 58.74 | $3.00 | Estimated-ranked (1 source family) |
| 78 | [Gemma 4 31B](/models/gemma-4-31b) | Google | Open Weight | 256K | 58.69 | Self-hosted / varies | Supported-ranked (2 source families) |
| 79 | [GPT-5.2 Instant](/models/gpt-5-2-instant) | OpenAI | Proprietary | 128K | 58.41 | $6.00 | Estimated-ranked (1 source family) |
| 80 | [Qwen3.5-27B](/models/qwen3-5-27b) | Alibaba | Open Weight | 262K | 58.32 | Self-hosted / varies | Supported-ranked (2 source families) |
| 81 | [GPT-5.3 Instant](/models/gpt-5-3-instant) | OpenAI | Proprietary | 400K | 58.3 | $14.00 | Estimated-ranked (1 source family) |
| 82 | [Grok 4.1 Fast (Reasoning)](/models/grok-4-1-fast-reasoning) | xAI | Proprietary | 2M | 58.18 | Not listed | Supported-ranked (2 source families) |
| 83 | [GPT-5 (high)](/models/gpt-5-high) | OpenAI | Proprietary | 128K | 58.04 | $10.00 | Estimated-ranked (1 source family) |
| 84 | [Gemini 2.5 Pro](/models/gemini-2-5-pro) | Google | Proprietary | 1M | 57.98 | $10.00 | Supported-ranked (3 source families) |
| 85 | [DeepSeek V3.2](/models/deepseek-v3-2) | DeepSeek | Open Weight | 128K | 57.62 | $0.42 | Supported-ranked (3 source families) |
| 86 | [DeepSeek V3.2 (Thinking)](/models/deepseek-v3-2-thinking) | DeepSeek | Open Weight | 128K | 57.54 | $2.19 | Estimated-ranked (1 source family) |
| 87 | [Qwen3 235B 2507 (Reasoning)](/models/qwen3-235b-2507-reasoning) | Alibaba | Open Weight | 128K | 57.43 | Self-hosted / varies | Estimated-ranked (1 source family) |
| 88 | [MiniMax M2.5](/models/minimax-m2-5) | MiniMax | Proprietary | 128K | 57.31 | $1.20 | Supported-ranked (3 source families) |
| 89 | [Hy3 Preview](/models/hy3-preview) | Tencent | Open Weight | 256K | 57.08 | Self-hosted / varies | Estimated-ranked (1 source family) |
| 90 | [GLM-4.5](/models/glm-4-5) | Z.AI | Proprietary | 128K | 56.99 | $2.20 | Estimated-ranked (1 source family) |
| 91 | [Gemma 4 26B A4B](/models/gemma-4-26b-a4b) | Google | Open Weight | 256K | 56.66 | Self-hosted / varies | Supported-ranked (3 source families) |
| 92 | [Qwen3.5 397B](/models/qwen3-5-397b) | Alibaba | Open Weight | 128K | 56.44 | $3.60 | Estimated-ranked (1 source family) |
| 93 | [GPT-5.2-Codex](/models/gpt-5-2-codex) | OpenAI | Proprietary | 400K | 56.38 | $14.00 | Supported-ranked (2 source families) |
| 94 | [GPT-5.3-Codex-Spark](/models/gpt-5-3-codex-spark) | OpenAI | Proprietary | 256K | 56.34 | Not listed | Estimated-ranked (1 source family) |
| 95 | [Gemini 3.1 Flash-Lite](/models/gemini-3-1-flash-lite) | Google | Proprietary | 1M | 56.13 | $1.50 | Supported-ranked (2 source families) |
| 96 | [Claude Opus 4.5 Thinking](/models/claude-opus-4-5-thinking) | Anthropic | Proprietary | 200K | 56.09 | Not listed | Estimated-ranked (1 source family) |
| 97 | [Qwen3.5 Flash](/models/qwen3-5-flash) | Alibaba | Proprietary | 1M | 56.03 | $0.40 | Supported-ranked (2 source families) |
| 98 | [Kimi K2.5](/models/kimi-k2-5) | Moonshot AI | Open Weight | 256K | 55.63 | $3.00 | Supported-ranked (3 source families) |
| 99 | [Qwen3 235B 2507](/models/qwen3-235b-2507) | Alibaba | Open Weight | 128K | 55.47 | Self-hosted / varies | Estimated-ranked (1 source family) |
| 100 | [Trinity-Large-Preview](/models/trinity-large-preview) | Arcee AI | Open Weight | 512K | 55.39 | $1.00 | Estimated-ranked (1 source family) |
| 101 | [Grok 4](/models/grok-4) | xAI | Proprietary | 128K | 54.91 | Not listed | Supported-ranked (3 source families) |
| 102 | [Apodex 1.1](/models/apodex-1-1) | Apodex | Proprietary | N/A | 54.9 | Not listed | Estimated-ranked (1 source family) |
| 103 | [Qwen3.5-35B-A3B](/models/qwen3-5-35b-a3b) | Alibaba | Open Weight | 262K | 54.86 | Self-hosted / varies | Supported-ranked (3 source families) |
| 104 | [Grok 4 Fast (Reasoning)](/models/grok-4-fast-reasoning) | xAI | Proprietary | 2M | 54.56 | Not listed | Supported-ranked (2 source families) |
| 105 | [DeepSeek LLM 2.0](/models/deepseek-llm-2-0) | DeepSeek | Open Weight | 128K | 54.01 | Self-hosted / varies | Estimated-ranked (1 source family) |
| 106 | [GPT-5.1-Codex-Max](/models/gpt-5-1-codex-max) | OpenAI | Proprietary | 400K | 53.96 | $10.00 | Estimated-ranked (1 source family) |
| 107 | [GLM-4.6](/models/glm-4-6) | Z.AI | Open Weight | 200K | 53.94 | Not listed | Supported-ranked (3 source families) |
| 108 | [Claude Sonnet 4.5](/models/claude-sonnet-4-5) | Anthropic | Proprietary | 200K | 53.74 | $15.00 | Supported-ranked (2 source families) |
| 109 | [Step 3.7 Flash](/models/step-3-7-flash) | StepFun | Open Weight | 256K | 53.19 | $1.15 | Estimated-ranked (1 source family) |
| 110 | [Step 3.5 Flash](/models/step-3-5-flash) | StepFun | Open Weight | 256K | 53.09 | $0.30 | Supported-ranked (2 source families) |
| 111 | [GPT-5 (medium)](/models/gpt-5-medium) | OpenAI | Proprietary | 128K | 52.81 | Not listed | Supported-ranked (2 source families) |
| 112 | [Claude Haiku 4.5](/models/claude-haiku-4-5) | Anthropic | Proprietary | 200K | 52.73 | $5.00 | Supported-ranked (2 source families) |
| 113 | [Qwen3.6-27B](/models/qwen3-6-27b) | Alibaba | Open Weight | 262K | 52.73 | Self-hosted / varies | Estimated-ranked (2 source families) |
| 114 | [DeepSeek V3.1](/models/deepseek-v3-1) | DeepSeek | Open Weight | 128K | 52.67 | Self-hosted / varies | Supported-ranked (3 source families) |
| 115 | [Nemotron 3 Nano 30B](/models/nemotron-3-nano-30b) | NVIDIA | Open Weight | 32K | 52.45 | Self-hosted / varies | Estimated-ranked (1 source family) |
| 116 | [MAI-Thinking-1](/models/mai-thinking-1) | Microsoft | Proprietary | 256K | 52.28 | Not listed | Estimated-ranked (6 source families) |
| 117 | [Ling 3.0 Flash](/models/ling-3-0-flash) | InclusionAI | Open Weight | 262K | 52.2 | Not listed | Estimated-ranked (1 source family) |
| 118 | [MiMo-V2-Flash](/models/mimo-v2-flash) | Xiaomi | Open Weight | 256K | 52.18 | Self-hosted / varies | Supported-ranked (2 source families) |
| 119 | [DeepSeek V3.1 (Reasoning)](/models/deepseek-v3-1-reasoning) | DeepSeek | Open Weight | 128K | 51.76 | Self-hosted / varies | Supported-ranked (2 source families) |
| 120 | [Gemini 2.5 Flash](/models/gemini-2-5-flash) | Google | Proprietary | 1M | 51.68 | $2.50 | Supported-ranked (3 source families) |
| 121 | [Grok 3 [Beta]](/models/grok-3-beta) | xAI | Proprietary | 128K | 51.66 | Not listed | Supported-ranked (2 source families) |
| 122 | [Qwen3.5 Plus](/models/qwen3-5-plus) | Alibaba | Proprietary | 1M | 51.64 | $2.40 | Estimated-ranked (1 source family) |
| 123 | [GPT-5.1-Codex](/models/gpt-5-1-codex) | OpenAI | Proprietary | 400K | 51.62 | $10.00 | Estimated-ranked (1 source family) |
| 124 | [DeepSeek-R1](/models/deepseek-r1) | DeepSeek | Open Weight | 128K | 51.33 | $2.19 | Supported-ranked (3 source families) |
| 125 | [Muse Glimmer 30B](/models/muse-glimmer-30b) | Meta | Open Weight | 131K | 51.04 | Self-hosted / varies | Estimated-ranked (1 source family) |
| 126 | [Qwen3.7 Flash](/models/qwen3-7-flash) | Alibaba | Proprietary | 1M | 50.76 | $0.13 | Estimated-ranked (1 source family) |
| 127 | [Nemotron 3 Super 120B A12B](/models/nemotron-3-super-120b-a12b) | NVIDIA | Open Weight | 256K | 50.52 | Self-hosted / varies | Estimated-ranked (1 source family) |
| 128 | [DeepSeek Coder 2.0](/models/deepseek-coder-2-0) | DeepSeek | Open Weight | 128K | 49.84 | Not listed | Estimated-ranked (1 source family) |
| 129 | [Grok 4.1 Fast](/models/grok-4-1-fast) | xAI | Proprietary | 1M | 49.8 | $0.50 | Supported-ranked (2 source families) |
| 130 | [Seed 1.6](/models/seed-1-6) | ByteDance | Proprietary | 256K | 49.79 | Not listed | Estimated-ranked (1 source family) |
| 131 | [Nemotron 3 Super 100B](/models/nemotron-3-super-100b) | NVIDIA | Open Weight | 1M | 49.65 | Self-hosted / varies | Estimated-ranked (1 source family) |
| 132 | [o4-mini (high)](/models/o4-mini-high) | OpenAI | Proprietary | 200K | 49.57 | Not listed | Estimated-ranked (1 source family) |
| 133 | [DeepSeekMath V2](/models/deepseekmath-v2) | DeepSeek | Open Weight | 128K | 49.45 | Self-hosted / varies | Estimated-ranked (1 source family) |
| 134 | [Qwen2.5-1M](/models/qwen2-5-1m) | Alibaba | Open Weight | 1M | 49.45 | Self-hosted / varies | Estimated-ranked (1 source family) |
| 135 | [GLM-4.7-Flash](/models/glm-4-7-flash) | Z.AI | Open Weight | 200K | 49.41 | Self-hosted / varies | Supported-ranked (2 source families) |
| 136 | [Seed-2.0-Lite](/models/seed-2-0-lite) | ByteDance | Proprietary | 256K | 49.41 | Not listed | Estimated-ranked (1 source family) |
| 137 | [GPT-OSS 120B](/models/gpt-oss-120b) | OpenAI | Open Weight | 128K | 49.26 | Self-hosted / varies | Supported-ranked (3 source families) |
| 138 | [o3-pro](/models/o3-pro) | OpenAI | Proprietary | 200K | 49.16 | $80.00 | Supported-ranked (3 source families) |
| 139 | [o3](/models/o3) | OpenAI | Proprietary | 200K | 49.13 | $8.00 | Supported-ranked (3 source families) |
| 140 | [Mistral Medium 3.5 128B](/models/mistral-medium-3-5-128b) | Mistral | Open Weight | 256K | 48.95 | $7.50 | Estimated-ranked (1 source family) |
| 141 | [Ministral 3 14B (Reasoning)](/models/ministral-3-14b-reasoning) | Mistral | Open Weight | 128K | 48.92 | $0.20 | Estimated-ranked (1 source family) |
| 142 | [Mistral Large 3](/models/mistral-large-3) | Mistral | Proprietary | 128K | 48.77 | $1.50 | Supported-ranked (2 source families) |
| 143 | [Qwen3.6-35B-A3B](/models/qwen3-6-35b-a3b) | Alibaba | Open Weight | 262K | 48.34 | Not listed | Estimated-ranked (2 source families) |
| 144 | [Grok 3 Mini](/models/grok-3-mini) | xAI | Proprietary | 128K | 48.27 | $0.50 | Estimated-ranked (1 source family) |
| 145 | [Aion-2.0](/models/aion-2-0) | Aion Labs | Proprietary | 128K | 48.25 | $1.60 | Estimated-ranked (1 source family) |
| 146 | [o3-mini](/models/o3-mini) | OpenAI | Proprietary | 200K | 48.06 | $4.40 | Supported-ranked (3 source families) |
| 147 | [o1](/models/o1) | OpenAI | Proprietary | 200K | 47.89 | $60.00 | Estimated-ranked (2 source families) |
| 148 | [o1-preview](/models/o1-preview) | OpenAI | Proprietary | 200K | 47.79 | $60.00 | Supported-ranked (3 source families) |
| 149 | [Mercury 2](/models/mercury-2) | Inception | Proprietary | 128K | 47.28 | $0.75 | Supported-ranked (2 source families) |
| 150 | [Trinity-Large-Thinking](/models/trinity-large-thinking) | Arcee AI | Open Weight | 512K | 47.12 | $0.90 | Supported-ranked (2 source families) |
| 151 | [Command A+](/models/command-a-plus) | Cohere | Open Weight | 128K | 46.87 | $10.00 | Estimated-ranked (1 source family) |
| 152 | [Mistral Small 4](/models/mistral-small-4) | Mistral | Open Weight | 256K | 46.82 | $0.60 | Supported-ranked (2 source families) |
| 153 | [K-Exaone](/models/k-exaone) | LG AI Research | Proprietary | 256K | 46.76 | Not listed | Estimated-ranked (1 source family) |
| 154 | [Gemma 4 12B](/models/gemma-4-12b) | Google | Open Weight | 256K | 46.65 | Not listed | Estimated-ranked (1 source family) |
| 155 | [GLM-4.5-Air](/models/glm-4-5-air) | Z.AI | Proprietary | 128K | 46.34 | $1.10 | Supported-ranked (2 source families) |
| 156 | [GPT-5 mini](/models/gpt-5-mini) | OpenAI | Proprietary | 128K | 46.29 | $2.00 | Supported-ranked (3 source families) |
| 157 | [o1-pro](/models/o1-pro) | OpenAI | Proprietary | 200K | 45.51 | $600.00 | Estimated-ranked (1 source family) |
| 158 | [Z-1](/models/z-1) | Z | Proprietary | 128K | 44.81 | Not listed | Estimated-ranked (1 source family) |
| 159 | [Seed 1.6 Flash](/models/seed-1-6-flash) | ByteDance | Proprietary | 256K | 44.76 | Not listed | Estimated-ranked (1 source family) |
| 160 | [Mistral 8x7B](/models/mistral-8x7b) | Mistral | Open Weight | 32K | 44.66 | Self-hosted / varies | Estimated-ranked (1 source family) |
| 161 | [Moonshot v1](/models/moonshot-v1) | Moonshot AI | Proprietary | 128K | 44.47 | Not listed | Estimated-ranked (1 source family) |
| 162 | [Claude 4.1 Opus](/models/claude-4-1-opus) | Anthropic | Proprietary | 200K | 44.39 | $75.00 | Supported-ranked (2 source families) |
| 163 | [Seed-2.0-Mini](/models/seed-2-0-mini) | ByteDance | Proprietary | 256K | 44.28 | Not listed | Estimated-ranked (1 source family) |
| 164 | [Nemotron Ultra 253B](/models/nemotron-ultra-253b) | NVIDIA | Open Weight | 32K | 44.09 | Self-hosted / varies | Estimated-ranked (1 source family) |
| 165 | [Nemotron 3 Nano Omni 30B A3B](/models/nemotron-3-nano-omni-30b-a3b) | NVIDIA | Open Weight | 256K | 43.97 | Self-hosted / varies | Estimated-ranked (1 source family) |
| 166 | [Qwen3 Max](/models/qwen3-max) | Alibaba | Proprietary | 1M | 43.82 | Not listed | Estimated-ranked (2 source families) |
| 167 | [GPT-4.1](/models/gpt-4-1) | OpenAI | Proprietary | 1M | 43.82 | $8.00 | Supported-ranked (3 source families) |
| 168 | [DeepSeek V3](/models/deepseek-v3) | DeepSeek | Open Weight | 128K | 43.74 | $1.10 | Supported-ranked (3 source families) |
| 169 | [Ling 2.6 Flash](/models/ling-2-6-flash) | InclusionAI | Open Weight | 262K | 43.68 | Not listed | Estimated-ranked (1 source family) |
| 170 | [GPT-OSS 20B](/models/gpt-oss-20b) | OpenAI | Open Weight | 128K | 42.95 | Self-hosted / varies | Supported-ranked (3 source families) |
| 171 | [Gemma 4 E4B](/models/gemma-4-e4b) | Google | Open Weight | 128K | 42.92 | Self-hosted / varies | Estimated-ranked (1 source family) |
| 172 | [Sarvam 105B](/models/sarvam-105b) | Sarvam | Open Weight | 128K | 42.82 | Self-hosted / varies | Estimated-ranked (1 source family) |
| 173 | [LFM2.5-2.6B](/models/lfm2-5-2-6b) | LiquidAI | Open Weight | 128K | 42.48 | Self-hosted / varies | Estimated-ranked (1 source family) |
| 174 | [Granite 4.2 8B](/models/granite-4-2-8b) | IBM | Open Weight | 128K | 42.03 | Self-hosted / varies | Supported-ranked (2 source families) |
| 175 | [Gemma 4 E2B](/models/gemma-4-e2b) | Google | Open Weight | 128K | 41.94 | Self-hosted / varies | Estimated-ranked (1 source family) |
| 176 | [Claude 4 Sonnet](/models/claude-4-sonnet) | Anthropic | Proprietary | 200K | 41.77 | $15.00 | Supported-ranked (2 source families) |
| 177 | [LFM2.5-8B-A1B](/models/lfm2-5-8b-a1b) | LiquidAI | Open Weight | 128K | 41.42 | Self-hosted / varies | Estimated-ranked (1 source family) |
| 178 | [Solar Pro 2](/models/solar-pro-2) | Upstage | Proprietary | 128K | 41.21 | Not listed | Estimated-ranked (1 source family) |
| 179 | [Sarvam 30B](/models/sarvam-30b) | Sarvam | Open Weight | 64K | 40.78 | Self-hosted / varies | Estimated-ranked (1 source family) |
| 180 | [Nemotron 3 Ultra](/models/nemotron-3-ultra) | NVIDIA | Open Weight | 1M | 40.69 | Self-hosted / varies | Estimated-ranked (13 source families) |
| 181 | [Exaone 4.0 32B](/models/exaone-4-0-32b) | LG AI Research | Open Weight | 128K | 40.54 | Not listed | Estimated-ranked (1 source family) |
| 182 | [GPT-4o](/models/gpt-4o) | OpenAI | Proprietary | 128K | 40.49 | $10.00 | Supported-ranked (2 source families) |
| 183 | [Ministral 3 8B (Reasoning)](/models/ministral-3-8b-reasoning) | Mistral | Open Weight | 128K | 40.17 | $0.15 | Estimated-ranked (1 source family) |
| 184 | [GPT-5 nano](/models/gpt-5-nano) | OpenAI | Proprietary | 400K | 39.72 | $0.40 | Estimated-ranked (2 source families) |
| 185 | [Qwen2.5-VL-32B](/models/qwen2-5-vl-32b) | Alibaba | Open Weight | 32K | 39.65 | Self-hosted / varies | Estimated-ranked (1 source family) |
| 186 | [Llama 4 Behemoth](/models/llama-4-behemoth) | Meta | Open Weight | 32K | 39.6 | Self-hosted / varies | Estimated-ranked (1 source family) |
| 187 | [Ministral 3 3B (Reasoning)](/models/ministral-3-3b-reasoning) | Mistral | Open Weight | 128K | 39.32 | $0.10 | Estimated-ranked (1 source family) |
| 188 | [Exaone 4.0 1.2B](/models/exaone-4-0-1-2b) | LG AI Research | Open Weight | 128K | 39.3 | Not listed | Estimated-ranked (1 source family) |
| 189 | [Granite-4.0-H-1B](/models/granite-4-0-h-1b) | IBM | Open Weight | 128K | 39.25 | Self-hosted / varies | Estimated-ranked (1 source family) |
| 190 | [Granite-4.0-1B](/models/granite-4-0-1b) | IBM | Open Weight | 128K | 39.03 | Self-hosted / varies | Estimated-ranked (1 source family) |
| 191 | [Mistral 8x7B v0.2](/models/mistral-8x7b-v0-2) | Mistral | Open Weight | 32K | 38.95 | Self-hosted / varies | Estimated-ranked (1 source family) |
| 192 | [Granite-4.0-350M](/models/granite-4-0-350m) | IBM | Open Weight | 32K | 38.8 | Self-hosted / varies | Estimated-ranked (1 source family) |
| 193 | [Granite-4.0-H-350M](/models/granite-4-0-h-350m) | IBM | Open Weight | 32K | 38.8 | Self-hosted / varies | Estimated-ranked (1 source family) |
| 194 | [Gemma 3 27B](/models/gemma-3-27b) | Google | Open Weight | 32K | 38.2 | Self-hosted / varies | Supported-ranked (3 source families) |
| 195 | [Mistral Large 2](/models/mistral-large-2) | Mistral | Proprietary | 128K | 37.56 | Not listed | Supported-ranked (2 source families) |
| 196 | [Qwen2.5-72B](/models/qwen2-5-72b) | Alibaba | Open Weight | 128K | 37.32 | Self-hosted / varies | Supported-ranked (2 source families) |
| 197 | [Grok Code Fast 1](/models/grok-code-fast-1) | xAI | Proprietary | 256K | 37.21 | $1.50 | Supported-ranked (2 source families) |
| 198 | [Claude 3 Opus](/models/claude-3-opus) | Anthropic | Proprietary | 200K | 36.85 | $75.00 | Supported-ranked (3 source families) |
| 199 | [Ornith-1.5-35B-A3B](/models/ornith-1-5-35b-a3b) | Ornith AI | Open Weight | 262K | 36.52 | Self-hosted / varies | Estimated-ranked (8 source families) |
| 200 | [Llama 3.1 405B](/models/llama-3-1-405b) | Meta | Open Weight | 128K | 36.5 | Self-hosted / varies | Supported-ranked (2 source families) |
| 201 | [Llama 4 Scout](/models/llama-4-scout) | Meta | Open Weight | 10M | 35.64 | Self-hosted / varies | Supported-ranked (3 source families) |
| 202 | [Gemini 1.5 Pro](/models/gemini-1-5-pro) | Google | Proprietary | 2M | 34.89 | $5.00 | Supported-ranked (2 source families) |
| 203 | [Claude 4.1 Opus Thinking](/models/claude-4-1-opus-thinking) | Anthropic | Proprietary | 200K | 34.62 | Not listed | Supported-ranked (2 source families) |
| 204 | [GPT-4.1 mini](/models/gpt-4-1-mini) | OpenAI | Proprietary | 1M | 34.47 | $1.60 | Estimated-ranked (2 source families) |
| 205 | [GPT-4o mini](/models/gpt-4o-mini) | OpenAI | Proprietary | 128K | 34.22 | $0.60 | Supported-ranked (3 source families) |
| 206 | [DeepSeek R1 Distill Qwen 32B](/models/deepseek-r1-distill-qwen-32b) | DeepSeek | Open Weight | 128K | 34.18 | Self-hosted / varies | Estimated-ranked (2 source families) |
| 207 | [Qwen2.5 Coder 32B Instruct](/models/qwen2-5-coder-32b-instruct) | Alibaba | Open Weight | 128K | 34.03 | Self-hosted / varies | Supported-ranked (2 source families) |
| 208 | [Ministral 3 14B](/models/ministral-3-14b) | Mistral | Open Weight | 128K | 33.77 | $0.20 | Supported-ranked (2 source families) |
| 209 | [Phi-4](/models/phi-4) | Microsoft | Open Weight | 16K | 32.91 | Self-hosted / varies | Supported-ranked (3 source families) |
| 210 | [Mistral Medium 3](/models/mistral-medium-3) | Mistral | Proprietary | 128K | 32.53 | $2.00 | Estimated-ranked (2 source families) |
| 211 | [Claude 3.5 Sonnet](/models/claude-3-5-sonnet) | Anthropic | Proprietary | 200K | 31.93 | $15.00 | Supported-ranked (2 source families) |
| 212 | [Llama 3 70B](/models/llama-3-70b) | Meta | Open Weight | 128K | 30.52 | Self-hosted / varies | Supported-ranked (2 source families) |
| 213 | [Ornith-1.5-9B](/models/ornith-1-5-9b) | Ornith AI | Open Weight | 262K | 28.51 | Self-hosted / varies | Estimated-ranked (7 source families) |
| 214 | [GPT-4.1 nano](/models/gpt-4-1-nano) | OpenAI | Proprietary | 1M | 28.43 | $0.40 | Estimated-ranked (2 source families) |
| 215 | [Mixtral 8x22B Instruct v0.1](/models/mixtral-8x22b-instruct-v0-1) | Mistral | Open Weight | 64K | 26.61 | Self-hosted / varies | Supported-ranked (2 source families) |
| 216 | [Kimi K2](/models/kimi-k2) | Moonshot AI | Proprietary | 128K | 26.13 | $2.50 | Supported-ranked (2 source families) |
| 217 | [GPT-4 Turbo](/models/gpt-4-turbo) | OpenAI | Proprietary | 128K | 25.48 | $30.00 | Supported-ranked (3 source families) |
| 218 | [Llama 4 Maverick](/models/llama-4-maverick) | Meta | Open Weight | 1M | 24.2 | Self-hosted / varies | Supported-ranked (3 source families) |
| 219 | [MiniMax M1 80k](/models/minimax-m1-80k) | MiniMax | Proprietary | 80K | 24.19 | Not listed | Supported-ranked (2 source families) |
| 220 | [Nemotron 3.5 Lightning 30B A3B NVFP4](/models/nemotron-3-5-lightning-30b-a3b-nvfp4) | NVIDIA | Open Weight | 1M | 21.36 | Self-hosted / varies | Estimated-ranked (10 source families) |
| 221 | [Ministral 3 8B](/models/ministral-3-8b) | Mistral | Open Weight | 128K | 20.61 | $0.15 | Supported-ranked (2 source families) |
| 222 | [Nova Pro](/models/nova-pro) | Amazon | Proprietary | 128K | 19.95 | Not listed | Supported-ranked (2 source families) |
| 223 | [LFM2-24B-A2B](/models/lfm2-24b-a2b) | LiquidAI | Proprietary | 32K | 18.66 | Self-hosted / varies | Supported-ranked (2 source families) |
| 224 | [Ministral 3 3B](/models/ministral-3-3b) | Mistral | Open Weight | 128K | 18.27 | $0.10 | Supported-ranked (2 source families) |
| 225 | [Claude 3 Haiku](/models/claude-3-haiku) | Anthropic | Proprietary | 200K | 17.34 | $1.25 | Supported-ranked (3 source families) |
| 226 | [LFM2.5-1.2B-Thinking](/models/lfm2-5-1-2b-thinking) | LiquidAI | Proprietary | 32K | 16.18 | Self-hosted / varies | Supported-ranked (2 source families) |
| 227 | [Nemotron-4 15B](/models/nemotron-4-15b) | NVIDIA | Open Weight | 32K | 15.88 | Self-hosted / varies | Supported-ranked (2 source families) |
| 228 | [LFM2.5-1.2B-Instruct](/models/lfm2-5-1-2b-instruct) | LiquidAI | Proprietary | 32K | 15.49 | Self-hosted / varies | Supported-ranked (2 source families) |
| 229 | [Gemini 1.0 Pro](/models/gemini-1-0-pro) | Google | Proprietary | 32K | 15.31 | Not listed | Supported-ranked (3 source families) |
| 230 | [Mistral 7B v0.3](/models/mistral-7b-v0-3) | Mistral | Open Weight | 32K | 8.65 | Self-hosted / varies | Supported-ranked (2 source families) |
| 231 | [Laguna M.1](/models/laguna-m-1) | Poolside | Proprietary | 256K | 4.46 | Not listed | Estimated-ranked (5 source families) |
| 232 | [Laguna XS.2](/models/laguna-xs-2) | Poolside | Open Weight | 256K | 1.91 | Self-hosted / varies | Estimated-ranked (5 source families) |

## What changed in September 2026

- Claude Fable 5.1 (Anthropic) holds the #1 spot on the BenchAlign leaderboard at 82.95.
- 8 model releases tracked in September 2026.
- Google climbed 1 spot in the provider race.
- 130 Supported and 102 Estimated models appear in the public leaderboard.

## Explore more

- [About BenchLM](/about)
- [Models Directory](/models)
- [Benchmarks Directory](/benchmarks)
- [Compare AI Models](/compare)
- [Download the BenchLM Dataset](/data)
- [Scoring Methodology](/methodology)
- [AI Alternatives Directory](/alternatives)
- [BenchLM Blog](/blog)