↓ Skip to main content

Model Details

Every model in the OpenCode Go catalogue, in the same order as the tracker page. Benchmarks come from Artificial Analysis; pricing comes from Models.dev and the Go API, and request estimates from the public Go plan page. Back to the rankings.

Qwen4 Max
#

  • Model ID: qwen4-max
  • Family: qwen4-max
  • Status: active
  • Release date: 2026-11-16
  • Last updated (Models.dev): 2026-11-16
  • Context window: 1M
  • Max output tokens: 262,144
  • Reasoning: yes
  • Tool calling: yes
  • Attachment support: yes
  • Open weights: —

Benchmarks (Artificial Analysis):

Intelligence IndexValueTerminal-Bench 4.0FrontierCode 2.0SciCodeBlended $/1M
42.911.448.5%26.5%—$3.75

Performance (Artificial Analysis): 56 tok/s · 6.26 s to first token · 148 s to first answer token · 31.4 min per agentic task

Measured on AA’s Qwen4 Max variant against the vendor’s own API, so read it as relative — third-party hosts and OpenCode Go serve the same open-weight model at different speeds.

Pricing (per 1M tokens):

InputOutputCache ReadCache Write
$1.50$4.50$0.07$1.88

Estimated requests (Go plan): 463 per 5 hours

Available in: API Models.dev Docs

Hy4 flash Free
#

  • Model ID: hy4-flash-free
  • Family: Hy
  • Status: active
  • Release date: 2026-11-16
  • Last updated (Models.dev): 2026-11-16
  • Context window: 1M
  • Max output tokens: 65,536
  • Reasoning: yes
  • Tool calling: yes
  • Attachment support: yes
  • Open weights: —

Pricing (per 1M tokens):

InputOutputCache ReadCache Write
$0$0$0—

Estimated requests (Go plan): ∞ per 5 hours

Available in: API Models.dev Docs

Hy4
#

  • Model ID: hy4
  • Family: Hy
  • Status: active
  • Release date: 2026-11-04
  • Last updated (Models.dev): 2026-11-04
  • Context window: 1M
  • Max output tokens: 65,536
  • Reasoning: yes
  • Tool calling: yes
  • Attachment support: yes
  • Open weights: —

Benchmarks (Artificial Analysis):

Intelligence IndexValueTerminal-Bench 4.0FrontierCode 2.0SciCodeBlended $/1M
34.616.637.4%31.4%—$2.08

Performance (Artificial Analysis): 95 tok/s · 2.62 s to first token · 108 s to first answer token · 13.6 min per agentic task

Measured on AA’s Tencent Hy4 variant against the vendor’s own API, so read it as relative — third-party hosts and OpenCode Go serve the same open-weight model at different speeds.

Pricing (per 1M tokens):

InputOutputCache ReadCache Write
$0.83$2.50$0.04—

Estimated requests (Go plan): 2,710 per 5 hours

Available in: API Models.dev Docs

Crouton Alpha Free
#

  • Model ID: crouton-alpha-free
  • Family:
  • Status: deprecated
  • Release date: 2026-11-03
  • Last updated (Models.dev): 2026-11-03
  • Context window: 1M
  • Max output tokens: 270,000
  • Reasoning: yes
  • Tool calling: yes
  • Attachment support: yes
  • Open weights: —

Pricing (per 1M tokens):

InputOutputCache ReadCache Write
$0$0$0—

Available in: Models.dev

LongCat 2.5 Free
#

  • Model ID: longcat-2.5-free
  • Family: longcat
  • Status: active
  • Release date: 2026-11-02
  • Last updated (Models.dev): 2026-11-02
  • Context window: 1M
  • Max output tokens: 262,144
  • Reasoning: yes
  • Tool calling: yes
  • Attachment support: yes
  • Open weights: —

Pricing (per 1M tokens):

InputOutputCache ReadCache Write
$0$0$0—

Estimated requests (Go plan): ∞ per 5 hours

Available in: API Models.dev Docs

Iceman Contributor Free
#

  • Model ID: iceman-contributor-free
  • Family:
  • Status: active
  • Release date: 2026-10-24
  • Last updated (Models.dev): 2026-10-24
  • Context window: 262K
  • Max output tokens: 131,072
  • Reasoning: yes
  • Tool calling: yes
  • Attachment support: —
  • Open weights: —

Pricing (per 1M tokens):

InputOutputCache ReadCache Write
$0$0$0—

Estimated requests (Go plan): ∞ per 5 hours

Available in: API Models.dev Docs

Kimi K3.1 Pro
#

  • Model ID: kimi-k3.1-pro
  • Family: kimi-k3
  • Status: active
  • Release date: 2026-10-23
  • Last updated (Models.dev): 2026-10-23
  • Context window: 1M
  • Max output tokens: 1,048,576
  • Reasoning: yes
  • Tool calling: yes
  • Attachment support: yes
  • Open weights: yes

Benchmarks (Artificial Analysis):

Intelligence IndexValueTerminal-Bench 4.0FrontierCode 2.0SciCodeBlended $/1M
46.55.867.2%25.5%60.8%$8.00

Performance (Artificial Analysis): 31 tok/s · 3.56 s to first token · 159 s to first answer token · 39.3 min per agentic task

Measured on AA’s Kimi K3.1 Pro variant against the vendor’s own API, so read it as relative — third-party hosts and OpenCode Go serve the same open-weight model at different speeds.

Pricing (per 1M tokens):

InputOutputCache ReadCache Write
$2.00$10.00$0.20$2.00

Estimated requests (Go plan): 147 per 5 hours

Available in: API Models.dev Docs

MiniMax M3.1 Flash Free
#

  • Model ID: minimax-m3.1-flash-free
  • Family: minimax
  • Status: deprecated
  • Release date: 2026-10-19
  • Last updated (Models.dev): 2026-10-19
  • Context window: 1M
  • Max output tokens: 512,000
  • Reasoning: yes
  • Tool calling: yes
  • Attachment support: yes
  • Open weights: yes

Pricing (per 1M tokens):

InputOutputCache ReadCache Write
$0$0$0—

Available in: Models.dev

MiniMax M3.1 Flash
#

  • Model ID: minimax-m3.1-flash
  • Family: minimax
  • Status: active
  • Release date: 2026-10-19
  • Last updated (Models.dev): 2026-10-19
  • Context window: 1M
  • Max output tokens: 512,000
  • Reasoning: yes
  • Tool calling: yes
  • Attachment support: yes
  • Open weights: yes

Benchmarks (Artificial Analysis):

Intelligence IndexValueTerminal-Bench 4.0FrontierCode 2.0SciCodeBlended $/1M
29.159.625.8%20.8%53.4%$0.49

Performance (Artificial Analysis): 128 tok/s · 1.31 s to first token · 53 s to first answer token · 17.9 min per agentic task

Measured on AA’s MiniMax-M3.1-Flash variant against the vendor’s own API, so read it as relative — third-party hosts and OpenCode Go serve the same open-weight model at different speeds.

Pricing (per 1M tokens):

InputOutputCache ReadCache Write
$0.15$0.60$0.01—

Estimated requests (Go plan): 3,880 per 5 hours

Available in: API Models.dev Docs

DeepSeek V4.1 Flash 1017
#

  • Model ID: deepseek-v4.1-flash
  • Family: deepseek-flash
  • Status: active
  • Release date: 2026-10-17
  • Last updated (Models.dev): 2026-10-17
  • Context window: 1M
  • Max output tokens: 384,000
  • Reasoning: yes
  • Tool calling: yes
  • Attachment support: yes
  • Open weights: yes

Benchmarks (Artificial Analysis):

Intelligence IndexValueTerminal-Bench 4.0FrontierCode 2.0SciCodeBlended $/1M
18.538.09.1%6.1%38.5%$0.49

Performance (Artificial Analysis): 146 tok/s · 1.20 s to first token · 1 s to first answer token · 7.4 min per agentic task

Measured on AA’s DeepSeek V4.1 Flash 1017 (Non-reasoning) variant against the vendor’s own API, so read it as relative — third-party hosts and OpenCode Go serve the same open-weight model at different speeds.

Pricing (per 1M tokens):

InputOutputCache ReadCache Write
$0.15$0.60$0.003—

Estimated requests (Go plan): 43,300 per 5 hours

Available in: API Models.dev Docs

Gemini 4 Argon Contributor
#

  • Model ID: gemini-4-argon-contributor
  • Family: gemini
  • Status: deprecated
  • Release date: 2026-10-13
  • Last updated (Models.dev): 2026-10-13
  • Context window: 1M
  • Max output tokens: 65,536
  • Reasoning: yes
  • Tool calling: yes
  • Attachment support: yes
  • Open weights: —

Pricing (per 1M tokens):

InputOutputCache ReadCache Write
$4.00$20.00$0.20—

Available in: Models.dev

Gemini 4 Argon
#

  • Model ID: gemini-4-argon
  • Family: gemini
  • Status: active
  • Release date: 2026-10-13
  • Last updated (Models.dev): 2026-10-13
  • Context window: 1M
  • Max output tokens: 65,536
  • Reasoning: yes
  • Tool calling: yes
  • Attachment support: yes
  • Open weights: —

Benchmarks (Artificial Analysis):

Intelligence IndexValueTerminal-Bench 4.0FrontierCode 2.0SciCodeBlended $/1M
47.66.055.1%25.0%58.9%$8.00

Performance (Artificial Analysis): 58 tok/s · 15.09 s to first token · 15 s to first answer token · 29.1 min per agentic task

Measured on AA’s Gemini 4 Argon (Medium) variant against the vendor’s own API, so read it as relative — third-party hosts and OpenCode Go serve the same open-weight model at different speeds.

Pricing (per 1M tokens):

InputOutputCache ReadCache Write
$2.00$10.00$0.10—

Estimated requests (Go plan): 566 per 5 hours

Available in: API Models.dev Docs

Cat Jam Free
#

  • Model ID: cat-jam-free
  • Family:
  • Status: deprecated
  • Release date: 2026-10-12
  • Last updated (Models.dev): 2026-10-12
  • Context window: 1M
  • Max output tokens: 131,072
  • Reasoning: yes
  • Tool calling: yes
  • Attachment support: —
  • Open weights: —

Pricing (per 1M tokens):

InputOutputCache ReadCache Write
$0$0$0—

Available in: Models.dev

Claude Haiku 5.5
#

  • Model ID: claude-haiku-5-5
  • Family: claude-haiku
  • Status: active
  • Release date: 2026-10-07
  • Last updated (Models.dev): 2026-10-07
  • Context window: 1M
  • Max output tokens: 128,000
  • Reasoning: yes
  • Tool calling: yes
  • Attachment support: yes
  • Open weights: —

Benchmarks (Artificial Analysis):

Intelligence IndexValueTerminal-Bench 4.0FrontierCode 2.0SciCodeBlended $/1M
29.974.721.7%23.2%48.7%$0.40

Performance (Artificial Analysis): 227 tok/s · 18.99 s to first token · 19 s to first answer token · 6.5 min per agentic task

Measured on AA’s Claude Haiku 5.5 (High) variant against the vendor’s own API, so read it as relative — third-party hosts and OpenCode Go serve the same open-weight model at different speeds.

Pricing (per 1M tokens):

InputOutputCache ReadCache Write
$0.10$0.50$0.01$0.12

Estimated requests (Go plan): 3,850 per 5 hours

Available in: API Models.dev Docs

Beam Free
#

  • Model ID: beam-free
  • Family: beam
  • Status: active
  • Release date: 2026-10-05
  • Last updated (Models.dev): 2026-10-16
  • Context window: 1M
  • Max output tokens: 131,072
  • Reasoning: yes
  • Tool calling: yes
  • Attachment support: —
  • Open weights: yes

Pricing (per 1M tokens):

InputOutputCache ReadCache Write
$0$0$0—

Estimated requests (Go plan): ∞ per 5 hours

Available in: API Models.dev Docs

LongCat 2.5 Preview Free
#

  • Model ID: longcat-2.5-preview-free
  • Family: longcat
  • Status: deprecated
  • Release date: 2026-09-25
  • Last updated (Models.dev): 2026-09-25
  • Context window: 1M
  • Max output tokens: 131,072
  • Reasoning: yes
  • Tool calling: yes
  • Attachment support: yes
  • Open weights: —

Pricing (per 1M tokens):

InputOutputCache ReadCache Write
$0$0$0—

Available in: Models.dev

Space Bunny
#

  • Model ID: space-bunny
  • Family:
  • Status: deprecated
  • Release date: 2026-09-23
  • Last updated (Models.dev): 2026-09-23
  • Context window: 1M
  • Max output tokens: 524,288
  • Reasoning: yes
  • Tool calling: yes
  • Attachment support: yes
  • Open weights: —

Pricing (per 1M tokens):

InputOutputCache ReadCache Write
$0.15$0.60$0.03$0

Available in: Models.dev

MiMo-V2.6-Pro
#

  • Model ID: mimo-v2.6-pro
  • Family: mimo
  • Status: active
  • Release date: 2026-09-22
  • Last updated (Models.dev): 2026-09-22
  • Context window: 1M
  • Max output tokens: 131,072
  • Reasoning: yes
  • Tool calling: yes
  • Attachment support: yes
  • Open weights: yes

Benchmarks (Artificial Analysis):

Intelligence IndexValueTerminal-Bench 4.0FrontierCode 2.0SciCodeBlended $/1M
29.939.334.8%24.0%60.9%$0.76

Performance (Artificial Analysis): 56 tok/s · 3.92 s to first token · 43 s to first answer token · 36.0 min per agentic task

Measured on AA’s MiMo-V2.6-Pro variant against the vendor’s own API, so read it as relative — third-party hosts and OpenCode Go serve the same open-weight model at different speeds.

Pricing (per 1M tokens):

InputOutputCache ReadCache Write
$0.43$0.87$0.0036—

Estimated requests (Go plan): 3,260 per 5 hours

Available in: API Models.dev Docs

MiMo-V2.6-Flash
#

  • Model ID: mimo-v2.6-flash
  • Family: mimo
  • Status: active
  • Release date: 2026-09-22
  • Last updated (Models.dev): 2026-09-22
  • Context window: 1M
  • Max output tokens: 131,072
  • Reasoning: yes
  • Tool calling: yes
  • Attachment support: yes
  • Open weights: yes

Benchmarks (Artificial Analysis):

Intelligence IndexValueTerminal-Bench 4.0FrontierCode 2.0SciCodeBlended $/1M
26.7109.122.7%21.6%51.3%$0.24

Performance (Artificial Analysis): 41 tok/s · 4.88 s to first token · 34 s to first answer token · 53.1 min per agentic task

Measured on AA’s MiMo-V2.6-Flash variant against the vendor’s own API, so read it as relative — third-party hosts and OpenCode Go serve the same open-weight model at different speeds.

Pricing (per 1M tokens):

InputOutputCache ReadCache Write
$0.14$0.28$0.0028—

Estimated requests (Go plan): 30,100 per 5 hours

Available in: API Models.dev Docs

GPT-6 Luna
#

  • Model ID: gpt-6-luna
  • Family: gpt-luna
  • Status: active
  • Release date: 2026-09-22
  • Last updated (Models.dev): 2026-09-22
  • Context window: 1M
  • Max output tokens: 128,000
  • Reasoning: yes
  • Tool calling: yes
  • Attachment support: yes
  • Open weights: —

Benchmarks (Artificial Analysis):

Intelligence IndexValueTerminal-Bench 4.0FrontierCode 2.0SciCodeBlended $/1M
23.558.68.1%18.4%51.7%$0.40

Performance (Artificial Analysis): 100 tok/s · 16.17 s to first token · 16 s to first answer token · 7.3 min per agentic task

Measured on AA’s GPT-6 Luna (Xhigh) variant against the vendor’s own API, so read it as relative — third-party hosts and OpenCode Go serve the same open-weight model at different speeds.

Pricing (per 1M tokens):

InputOutputCache ReadCache Write
$0.10$0.50$0.01$0.12

Estimated requests (Go plan): 4,230 per 5 hours

Available in: API Models.dev Docs

Grok 4.7
#

  • Model ID: grok-4.7
  • Family: grok
  • Status: active
  • Release date: 2026-09-21
  • Last updated (Models.dev): 2026-09-21
  • Context window: 500K
  • Max output tokens: 500,000
  • Reasoning: yes
  • Tool calling: yes
  • Attachment support: yes
  • Open weights: —

Benchmarks (Artificial Analysis):

Intelligence IndexValueTerminal-Bench 4.0FrontierCode 2.0SciCodeBlended $/1M
31.16.216.2%21.8%54.9%$5.00

Performance (Artificial Analysis): 68 tok/s · 14.63 s to first token · 15 s to first answer token · 10.5 min per agentic task

Measured on AA’s Grok 4.7 (Low) variant against the vendor’s own API, so read it as relative — third-party hosts and OpenCode Go serve the same open-weight model at different speeds.

Pricing (per 1M tokens):

InputOutputCache ReadCache Write
$2.00$6.00$0.50—

Estimated requests (Go plan): 169 per 5 hours

Available in: API Models.dev Docs

Step 5 Preview Free
#

  • Model ID: step-5-preview-free
  • Family:
  • Status: deprecated
  • Release date: 2026-09-16
  • Last updated (Models.dev): 2026-09-20
  • Context window: 1M
  • Max output tokens: 65,536
  • Reasoning: yes
  • Tool calling: yes
  • Attachment support: yes
  • Open weights: yes

Pricing (per 1M tokens):

InputOutputCache ReadCache Write
$0$0$0—

Available in: Models.dev

Step 5 Preview
#

  • Model ID: step-5-preview
  • Family:
  • Status: active
  • Release date: 2026-09-16
  • Last updated (Models.dev): 2026-09-20
  • Context window: 1M
  • Max output tokens: 65,536
  • Reasoning: yes
  • Tool calling: yes
  • Attachment support: yes
  • Open weights: yes

Benchmarks (Artificial Analysis):

Intelligence IndexValueTerminal-Bench 4.0FrontierCode 2.0SciCodeBlended $/1M
28.512.533.3%12.0%58.9%$2.27

Performance (Artificial Analysis): 156 tok/s · 0.87 s to first token · 20 s to first answer token · 12.1 min per agentic task

Measured on AA’s Step 5 Preview variant against the vendor’s own API, so read it as relative — third-party hosts and OpenCode Go serve the same open-weight model at different speeds.

Pricing (per 1M tokens):

InputOutputCache ReadCache Write
$1.00$2.70$0.10—

Estimated requests (Go plan): 426 per 5 hours

Available in: API Models.dev Docs

Muse Spark 1.3 Contributor
#

  • Model ID: muse-spark-1.3-contributor
  • Family: muse
  • Status: active
  • Release date: 2026-09-02
  • Last updated (Models.dev): 2026-09-02
  • Context window: 1M
  • Max output tokens: 131,072
  • Reasoning: yes
  • Tool calling: yes
  • Attachment support: yes
  • Open weights: —

Pricing (per 1M tokens):

InputOutputCache ReadCache Write
$0.10$0.20$0.002—

Estimated requests (Go plan): 45,300 per 5 hours

Available in: API Models.dev Docs

Hy4 preview
#

  • Model ID: hy4-preview
  • Family: Hy
  • Status: active
  • Release date: 2026-08-28
  • Last updated (Models.dev): 2026-08-28
  • Context window: 1M
  • Max output tokens: 64,000
  • Reasoning: yes
  • Tool calling: yes
  • Attachment support: —
  • Open weights: yes

Pricing (per 1M tokens):

InputOutputCache ReadCache Write
$0.83$2.50$0.04—

Estimated requests (Go plan): 1,350 per 5 hours

Available in: API Models.dev Docs

Qwen3.8 Flash
#

  • Model ID: qwen3.8-flash
  • Family: qwen
  • Status: active
  • Release date: 2026-08-26
  • Last updated (Models.dev): 2026-08-26
  • Context window: 1M
  • Max output tokens: 131,072
  • Reasoning: yes
  • Tool calling: yes
  • Attachment support: yes
  • Open weights: —

Pricing (per 1M tokens):

InputOutputCache ReadCache Write
$0.15$0.47$0.02$0.20

Estimated requests (Go plan): 5,400 per 5 hours

Available in: API Models.dev Docs

GLM-5.3-Flash
#

  • Model ID: glm-5.3-flash
  • Family: glm-flash
  • Status: active
  • Release date: 2026-08-26
  • Last updated (Models.dev): 2026-08-26
  • Context window: 1M
  • Max output tokens: 131,072
  • Reasoning: yes
  • Tool calling: yes
  • Attachment support: yes
  • Open weights: yes

Benchmarks (Artificial Analysis):

Intelligence IndexValueTerminal-Bench 4.0FrontierCode 2.0SciCodeBlended $/1M
27.867.432.8%17.6%51.6%$0.41

Performance (Artificial Analysis): 57 tok/s · 3.27 s to first token · 91 s to first answer token · 39.4 min per agentic task

Measured on AA’s GLM-5.3-Flash variant against the vendor’s own API, so read it as relative — third-party hosts and OpenCode Go serve the same open-weight model at different speeds.

Pricing (per 1M tokens):

InputOutputCache ReadCache Write
$0.15$0.50$0.03—

Estimated requests (Go plan): 6,320 per 5 hours

Available in: API Models.dev Docs

GLM-5.3
#

  • Model ID: glm-5.3
  • Family: glm
  • Status: active
  • Release date: 2026-08-14
  • Last updated (Models.dev): 2026-08-14
  • Context window: 1M
  • Max output tokens: 131,072
  • Reasoning: yes
  • Tool calling: yes
  • Attachment support: —
  • Open weights: yes

Benchmarks (Artificial Analysis):

Intelligence IndexValueTerminal-Bench 4.0FrontierCode 2.0SciCodeBlended $/1M
23.16.334.8%12.7%42.0%$3.65

Performance (Artificial Analysis): 65 tok/s · 2.94 s to first token · 10 s to first answer token · 13.4 min per agentic task

Measured on AA’s GLM-5.3 (Low) variant against the vendor’s own API, so read it as relative — third-party hosts and OpenCode Go serve the same open-weight model at different speeds.

Pricing (per 1M tokens):

InputOutputCache ReadCache Write
$1.40$4.40$0.26—

Estimated requests (Go plan): 6,600 per 5 hours

Available in: API Models.dev Docs

Muse Spark 1.2 Contributor
#

  • Model ID: muse-spark-1.2-contributor
  • Family: muse
  • Status: active
  • Release date: 2026-08-05
  • Last updated (Models.dev): 2026-08-05
  • Context window: 1M
  • Max output tokens: 131,072
  • Reasoning: yes
  • Tool calling: yes
  • Attachment support: yes
  • Open weights: —

Pricing (per 1M tokens):

InputOutputCache ReadCache Write
$0.10$0.20$0.002—

Estimated requests (Go plan): 45,300 per 5 hours

Available in: API Models.dev Docs

Qwen3.8 Max
#

  • Model ID: qwen3.8-max
  • Family: qwen3.8-max
  • Status: active
  • Release date: 2026-08-03
  • Last updated (Models.dev): 2026-08-03
  • Context window: 1M
  • Max output tokens: 131,072
  • Reasoning: yes
  • Tool calling: yes
  • Attachment support: yes
  • Open weights: —

Pricing (per 1M tokens):

InputOutputCache ReadCache Write
$2.00$6.00$0.25$2.50

Estimated requests (Go plan): 162 per 5 hours

Available in: API Models.dev Docs

Kimi K3
#

  • Model ID: kimi-k3
  • Family: kimi-k3
  • Status: active
  • Release date: 2026-07-16
  • Last updated (Models.dev): 2026-07-16
  • Context window: 1M
  • Max output tokens: 1,048,576
  • Reasoning: yes
  • Tool calling: yes
  • Attachment support: yes
  • Open weights: yes

Pricing (per 1M tokens):

InputOutputCache ReadCache Write
$3.00$15.00$0.30$3.00

Estimated requests (Go plan): 98 per 5 hours

Available in: API Models.dev Docs

GPT-5.6 Luna
#

  • Model ID: gpt-5.6-luna
  • Family: gpt-luna
  • Status: active
  • Release date: 2026-07-09
  • Last updated (Models.dev): 2026-07-09
  • Context window: 1M
  • Max output tokens: 128,000
  • Reasoning: yes
  • Tool calling: yes
  • Attachment support: yes
  • Open weights: —

Pricing (per 1M tokens):

InputOutputCache ReadCache Write
$0.20$1.20$0.02$0.25

Estimated requests (Go plan): 2,050 per 5 hours

Available in: API Models.dev Docs

Hy3
#

  • Model ID: hy3
  • Family: Hy
  • Status: active
  • Release date: 2026-07-06
  • Last updated (Models.dev): 2026-07-06
  • Context window: 256K
  • Max output tokens: 128,000
  • Reasoning: yes
  • Tool calling: yes
  • Attachment support: —
  • Open weights: yes

Pricing (per 1M tokens):

InputOutputCache ReadCache Write
$0.14$0.58$0.04—

Estimated requests (Go plan): 4,300 per 5 hours

Available in: API Models.dev Docs

LongCat-2.0
#

  • Model ID: longcat-2.0
  • Family: longcat
  • Status: active
  • Release date: 2026-06-30
  • Last updated (Models.dev): 2026-06-30
  • Context window: 1M
  • Max output tokens: 131,072
  • Reasoning: yes
  • Tool calling: yes
  • Attachment support: —
  • Open weights: yes

Performance (Artificial Analysis): 58 tok/s · 3.26 s to first token · 49 s to first answer token

Measured on AA’s LongCat 2.0 variant against the vendor’s own API, so read it as relative — third-party hosts and OpenCode Go serve the same open-weight model at different speeds.

Pricing (per 1M tokens):

InputOutputCache ReadCache Write
$0.30$1.20$0.006—

Estimated requests (Go plan): 11,400 per 5 hours

Available in: API Models.dev Docs

GLM-5.2
#

  • Model ID: glm-5.2
  • Family: glm
  • Status: deprecated
  • Release date: 2026-06-13
  • Last updated (Models.dev): 2026-06-13
  • Context window: 1M
  • Max output tokens: 131,072
  • Reasoning: yes
  • Tool calling: yes
  • Attachment support: —
  • Open weights: yes

Pricing (per 1M tokens):

InputOutputCache ReadCache Write
$1.40$4.40$0.26—

Estimated requests (Go plan): 792 per 5 hours

Available in: Models.dev Docs

Kimi K2.7 Code
#

  • Model ID: kimi-k2.7-code
  • Family: kimi-k2
  • Status: active
  • Release date: 2026-06-12
  • Last updated (Models.dev): 2026-06-12
  • Context window: 262K
  • Max output tokens: 262,144
  • Reasoning: yes
  • Tool calling: yes
  • Attachment support: yes
  • Open weights: yes

Performance (Artificial Analysis): 56 tok/s · 6.26 s to first token · 34 s to first answer token

Measured on AA’s Kimi K2.7 Code variant against the vendor’s own API, so read it as relative — third-party hosts and OpenCode Go serve the same open-weight model at different speeds.

Pricing (per 1M tokens):

InputOutputCache ReadCache Write
$0.95$4.00$0.19—

Estimated requests (Go plan): 994 per 5 hours

Available in: API Models.dev Docs

Qwen3.7 Plus
#

  • Model ID: qwen3.7-plus
  • Family: qwen3.7-plus
  • Status: active
  • Release date: 2026-06-02
  • Last updated (Models.dev): 2026-06-02
  • Context window: 1M
  • Max output tokens: 65,536
  • Reasoning: yes
  • Tool calling: yes
  • Attachment support: yes
  • Open weights: —

Performance (Artificial Analysis): 56 tok/s · 2.77 s to first token · 47 s to first answer token

Measured on AA’s Qwen3.7 Plus variant against the vendor’s own API, so read it as relative — third-party hosts and OpenCode Go serve the same open-weight model at different speeds.

Pricing (per 1M tokens):

InputOutputCache ReadCache Write
$0.40$1.60$0.04$0.50

Estimated requests (Go plan): 4,310 per 5 hours

Available in: API Models.dev Docs

MiniMax-M3
#

  • Model ID: minimax-m3
  • Family: minimax-m3
  • Status: active
  • Release date: 2026-06-01
  • Last updated (Models.dev): 2026-06-01
  • Context window: 1M
  • Max output tokens: 131,072
  • Reasoning: yes
  • Tool calling: yes
  • Attachment support: yes
  • Open weights: yes

Benchmarks (Artificial Analysis):

Intelligence IndexValueTerminal-Bench 4.0FrontierCode 2.0SciCodeBlended $/1M
18.919.42.0%21.9%47.1%$0.97

Performance (Artificial Analysis): 91 tok/s · 2.32 s to first token · 91 s to first answer token · 16.7 min per agentic task

Measured on AA’s MiniMax-M3 variant against the vendor’s own API, so read it as relative — third-party hosts and OpenCode Go serve the same open-weight model at different speeds.

Pricing (per 1M tokens):

InputOutputCache ReadCache Write
$0.30$1.20$0.06—

Estimated requests (Go plan): 3,210 per 5 hours

Available in: API Models.dev Docs

DeepSeek V4 Pro
#

  • Model ID: deepseek-v4-pro
  • Family: deepseek-thinking
  • Status: active
  • Release date: 2026-04-24
  • Last updated (Models.dev): 2026-04-24
  • Context window: 1M
  • Max output tokens: 384,000
  • Reasoning: yes
  • Tool calling: yes
  • Attachment support: —
  • Open weights: yes

Benchmarks (Artificial Analysis):

Intelligence IndexValueTerminal-Bench 4.0FrontierCode 2.0SciCodeBlended $/1M
23.114.014.1%17.1%51.0%$1.65

Performance (Artificial Analysis): 87 tok/s · 2.43 s to first token · 57 s to first answer token · 18.9 min per agentic task

Measured on AA’s DeepSeek V4 Pro 0813 (Max) variant against the vendor’s own API, so read it as relative — third-party hosts and OpenCode Go serve the same open-weight model at different speeds.

Pricing (per 1M tokens):

InputOutputCache ReadCache Write
$0.66$1.98$0.02—

Estimated requests (Go plan): 1,040 per 5 hours

Available in: API Models.dev Docs

MiMo V2.5 Pro
#

  • Model ID: mimo-v2.5-pro
  • Family: mimo-v2.5-pro
  • Status: active
  • Release date: 2026-04-22
  • Last updated (Models.dev): 2026-04-22
  • Context window: 1M
  • Max output tokens: 128,000
  • Reasoning: yes
  • Tool calling: yes
  • Attachment support: —
  • Open weights: yes

Pricing (per 1M tokens):

InputOutputCache ReadCache Write
$0.43$0.87$0.0036—

Available in: Models.dev

MiMo V2.5
#

  • Model ID: mimo-v2.5
  • Family: mimo-v2.5
  • Status: active
  • Release date: 2026-04-22
  • Last updated (Models.dev): 2026-04-22
  • Context window: 1M
  • Max output tokens: 128,000
  • Reasoning: yes
  • Tool calling: yes
  • Attachment support: yes
  • Open weights: yes

Pricing (per 1M tokens):

InputOutputCache ReadCache Write
$0.14$0.28$0.0028—

Available in: Models.dev

MiniMax-M2.7
#

  • Model ID: minimax-m2.7
  • Family: minimax
  • Status: deprecated
  • Release date: 2026-03-18
  • Last updated (Models.dev): 2026-03-18
  • Context window: 205K
  • Max output tokens: 131,072
  • Reasoning: yes
  • Tool calling: yes
  • Attachment support: —
  • Open weights: yes

Pricing (per 1M tokens):

InputOutputCache ReadCache Write
$0.30$1.20$0.06$0.38

Available in: Models.dev