HealthBench Professional

Zhipu logoGLM 5.3 on HealthBench Professional

Rank 29 of 43, updated November 2, 2026.

GLM 5.3 scores 0.528 on HealthBench Professional, rank 29 of 43 models on the board. HealthBench Professional scores models on 525 tasks drawn from real clinician conversations, graded criterion by criterion against physician-written rubrics on a 0 to 1 scale.

29
rank of 43
0.528
length-adjusted score
0.579
score before the length adjustment
0.175
gap to the leader, GPT-6 Astra (Anthropic run)
1M
context window, tokens
$1.40 / $4.40
input / output price per 1M tokens

Where it sits on the board

The models ranked around GLM 5.3, by length-adjusted score. Differences are against GLM 5.3.

RankModelLength-adjustedBefore the adjustmentDifference
26Meta logoMuse Spark0.541–+0.013
27OpenAI logoGPT-5.6 Sol (August)0.5400.566+0.012
28Moonshot AI logoKimi K30.5320.569+0.004
29Zhipu logoGLM 5.30.5280.579–
30Anthropic logoClaude Opus 4.70.519–-0.009
31OpenAI logoGPT-5.50.5180.572-0.010
32Google logoGemini 3.8 Flash0.5170.590-0.011

Result and API facts

Rank29 of 43
Length-adjusted score0.528
Score before the length adjustment0.579
LabZhipu
Context window1M tokens
API price per 1M tokens$1.40 in / $4.40 out
Licenseopen
SourceHealthBench Professional Benchmark | Artificial Analysis (independent leaderboard)
Released2026-08-18

Source of this score

Read from HealthBench Professional Benchmark | Artificial Analysis (independent leaderboard, Artificial Analysis, 2026-11-02). Independent run. Confidence: partial. Configuration: Artificial Analysis independent run, max reasoning; GPT-5.4 (low) judge, 4 repeats per task; length-adjusted (57.9 raw, 3,748 mean response chars).

Score section, Length-Adjusted Score chart, GLM-5.3 (max) bar label; raw score 57.9% on the Raw Score tabfull entry on the sources page
HealthBench Professional: Length-Adjusted Score | GLM-5.3 (max) | 52.8%

Position in the field

The gap to the leader, GPT-6 Astra (Anthropic run) at 0.703, is 0.175. Directly above sits Kimi K3 at 0.532. Directly below sits Claude Opus 4.7 at 0.519. The scores on this page are compiled from published documents rather than from one controlled run, so a small gap between 2 models can reflect a difference in grader version, reasoning effort, or deployment setting as well as a difference in capability.

What does GLM 5.3 score on HealthBench Professional?

GLM 5.3 scores 0.528 on HealthBench Professional, which places it at rank 29 of 43 models on the board as of November 2, 2026. The number was read from HealthBench Professional Benchmark | Artificial Analysis, listed on the sources page.

How much does GLM 5.3 cost to run?

GLM 5.3 is priced at $1.40 per million input tokens and $4.40 per million output tokens through Zhipu's API.

Pairings with a dedicated comparison page are linked; every other difference is in the score-difference matrix.

  • GLM 5.3 vs GPT-6 Astra (Anthropic run)
    0.528 vs 0.703, GPT-6 Astra (Anthropic run) by 0.175
  • GLM 5.3 vs Claude Sonnet 5.5
    0.528 vs 0.692, Claude Sonnet 5.5 by 0.164
  • GLM 5.3 vs Claude Fable 5
    0.528 vs 0.660, Claude Fable 5 by 0.132
  • GLM 5.3 vs Claude Opus 5.5
    0.528 vs 0.656, Claude Opus 5.5 by 0.128
  • GLM 5.3 vs GPT-6 Luna (Anthropic run)
    0.528 vs 0.655, GPT-6 Luna (Anthropic run) by 0.127
  • GLM 5.3 vs Claude Haiku 5.5
    0.528 vs 0.648, Claude Haiku 5.5 by 0.120
  • GLM 5.3 vs GPT-6 Astra
    0.528 vs 0.647, GPT-6 Astra by 0.119
  • GLM 5.3 vs GPT-6.1 Sol
    0.528 vs 0.642, GPT-6.1 Sol by 0.114
  • GLM 5.3 vs Claude Fable 5 (September card)
    0.528 vs 0.633, Claude Fable 5 (September card) by 0.105
  • GLM 5.3 vs GPT-6.1 Terra
    0.528 vs 0.627, GPT-6.1 Terra by 0.099
  • GLM 5.3 vs Muse Wonder Full
    0.528 vs 0.623, Muse Wonder Full by 0.095
  • GLM 5.3 vs Claude Fable 5.1
    0.528 vs 0.621, Claude Fable 5.1 by 0.093
  • GLM 5.3 vs GPT-6 Luna
    0.528 vs 0.608, GPT-6 Luna by 0.080
  • GLM 5.3 vs GPT-6 Sol
    0.528 vs 0.608, GPT-6 Sol by 0.080
  • GLM 5.3 vs GPT-5.6 Sol
    0.528 vs 0.605, GPT-5.6 Sol by 0.077
  • GLM 5.3 vs Muse Wonder
    0.528 vs 0.605, Muse Wonder by 0.077
  • GLM 5.3 vs Claude Opus 5
    0.528 vs 0.598, Claude Opus 5 by 0.070
  • GLM 5.3 vs Muse Spark 1.1
    0.528 vs 0.593, Muse Spark 1.1 by 0.065
  • GLM 5.3 vs Claude Sonnet 5
    0.528 vs 0.578, Claude Sonnet 5 by 0.050
  • GLM 5.3 vs GPT-5.6 Terra
    0.528 vs 0.577, GPT-5.6 Terra by 0.049
  • GLM 5.3 vs Claude Opus 4.8 (Opus 4.8 grader)
    0.528 vs 0.574, Claude Opus 4.8 (Opus 4.8 grader) by 0.046
  • GLM 5.3 vs Grok 4.7
    0.528 vs 0.567, Grok 4.7 by 0.039
  • GLM 5.3 vs Claude Opus 4.8
    0.528 vs 0.558, Claude Opus 4.8 by 0.030
  • GLM 5.3 vs GPT-5.6 Luna
    0.528 vs 0.557, GPT-5.6 Luna by 0.029
  • GLM 5.3 vs Gemini 4 Argon
    0.528 vs 0.551, Gemini 4 Argon by 0.023
  • GLM 5.3 vs Muse Spark
    0.528 vs 0.541, Muse Spark by 0.013
  • GLM 5.3 vs GPT-5.6 Sol (August)
    0.528 vs 0.540, GPT-5.6 Sol (August) by 0.012
  • GLM 5.3 vs Kimi K3
    0.528 vs 0.532, Kimi K3 by 0.004
  • GLM 5.3 vs Claude Opus 4.7
    0.528 vs 0.519, GLM 5.3 by 0.009
  • GLM 5.3 vs GPT-5.5
    0.528 vs 0.518, GLM 5.3 by 0.010
  • GLM 5.3 vs Gemini 3.8 Flash
    0.528 vs 0.517, GLM 5.3 by 0.011
  • GLM 5.3 vs DeepSeek V4.1 Flash
    0.528 vs 0.507, GLM 5.3 by 0.021
  • GLM 5.3 vs Grok 4.6
    0.528 vs 0.485, GLM 5.3 by 0.043
  • GLM 5.3 vs GPT-5.4
    0.528 vs 0.481, GLM 5.3 by 0.047
  • GLM 5.3 vs GPT-5
    0.528 vs 0.462, GLM 5.3 by 0.066
  • GLM 5.3 vs GPT-5.2
    0.528 vs 0.459, GLM 5.3 by 0.069
  • GLM 5.3 vs Claude Sonnet 4.6
    0.528 vs 0.442, GLM 5.3 by 0.086
  • GLM 5.3 vs GPT-5.6 Luna (August)
    0.528 vs 0.441, GLM 5.3 by 0.087
  • GLM 5.3 vs GPT-5.1
    0.528 vs 0.396, GLM 5.3 by 0.132
  • GLM 5.3 vs GPT-5.5 Instant
    0.528 vs 0.384, GLM 5.3 by 0.144
  • GLM 5.3 vs MAI-Thinking-1
    0.528 vs 0.350, GLM 5.3 by 0.178
  • GLM 5.3 vs Claude Haiku 4.5
    0.528 vs 0.322, GLM 5.3 by 0.206

Where the scores come from and how they are read is on the methodology page. The full ranking is on the leaderboard.