Amd

4 across 1 category board · 15 without a live endpoint · 63 equivalent routes · 5 superseded

Language ModelCap Index

Language rank	Model	ModelCap Index	Gravity	Context	Input tokens	Output tokens	Released
#343	HyLo Llama 14MLA14GDN 64K SFTOpen weights	6.9	4.2	—	—	—	last month
#344	HyLo Llama 6MLA22GDN 64K SFTOpen weights	6.9	4.2	—	—	—	last month
#351	HyLo Llama 4MLA12GDN 64K SFTOpen weights	3.4	4.2	—	—	—	last month
#352	HyLo Llama 8MLA8GDN 64K SFTOpen weights	3.4	4.5	—	—	—	last month

No live endpoint

Board status	Model	Gravity	Context	Input tokens	Output tokens	Released
No live endpoint	Instella MoE 16B A3B ThinkRestricted license Language	10.8	—	—	—	3 months ago
No live endpoint	tiny qwen3 moe w4a8Open weights Language	3.6	—	—	—	4 months ago
No live endpoint	PARD2 Llama 3.1 8BOpen weights Language	11.0	—	—	—	5 months ago
No live endpoint	PARD2 Qwen3 8BOpen weights Language	12.1	—	—	—	5 months ago
No live endpoint	Instella 3B MathRestricted license Language	8.4	—	—	—	last year
No live endpoint	PARD Qwen3 0.6BOpen weights Language	9.0	—	—	—	last year
No live endpoint	PARD Llama 4 1BOpen weights Language	6.0	—	—	—	last year
No live endpoint	Instella 3B Long InstructRestricted license Language	8.2	—	—	—	last year
No live endpoint	PARD DeepSeek R1 Distill Qwen 1.5BOpen weights Language	1.5	—	—	—	last year
No live endpoint	PARD Qwen2.5 0.5BOpen weights Language	6.3	—	—	—	last year
No live endpoint	Instella 3B instructRestricted license Language	8.0	—	—	—	last year
No live endpoint	Instella 3B Stage1Restricted license Language	8.0	—	—	—	last year
No live endpoint	AMD OLMo 1BOpen weights Language	7.2	—	—	—	2 years ago
No live endpoint	AMD Llama 135m codeOpen weights Language	0.7	—	—	—	2 years ago
No live endpoint	AMD Llama 135mOpen weights Language	11.0	—	—	—	2 years ago

Reviewed equivalent routes

Board status	Model	Gravity	Context	Input tokens	Output tokens	Released
Equivalent route	Qwen3.5 397B A17B MXFP4 AttnFP8 V3Open weights Language	7.0	—	—	—	last month
Equivalent route	Qwen3.6 35B A3B w4a16 llmcompressorOpen weights Language	11.3	—	—	—	last month
Equivalent route	Instella MoE 16B A3B SFT w4a16 llmcompressorRestricted license Language	11.7	—	—	—	last month
Equivalent route	Qwen3.8 Flash Next Quark MXFP4 PLEFP8Restricted license Language	11.6	—	—	—	last month
Equivalent route	Qwen3.8 27B w8a8 llmcompressorOpen weights Language	8.3	—	—	—	last month
Equivalent route	Instella MoE 16B A3B SFT w8a8 llmcompressorRestricted license Language	11.6	—	—	—	last month
Equivalent route	DeepSeek V4.1 Flash Quark MXFP4Open weights Language	8.7	—	—	—	last month
Equivalent route	GLM 5.3 Quark MXFP4 AttnFP8Restricted license Language	11.4	—	—	—	2 months ago
Equivalent route	Qwen3.8 Flash Next Quark MXFP4Restricted license Language	11.3	—	—	—	2 months ago
Equivalent route	gemma 4 26B A4B it w4a16 llmcompressorRestricted license Language	11.3	—	—	—	2 months ago
Equivalent route	gemma 4 12B it w4a16 llmcompressorRestricted license Language	11.2	—	—	—	2 months ago
Equivalent route	Qwen3.8 27B Quark AWQ MXFP4Open weights Language	16.3	—	—	—	2 months ago
Represented by Qwen3.8 27B Quark AWQ MXFP4	Qwen3.8 27B Quark AWQ INT4 W4A16Open weights Language	14.5	—	—	—	2 months ago
Equivalent route	Qwen3.8 27B Quark Qronos INT4 W4A16Open weights Language	7.4	—	—	—	2 months ago
Equivalent route	Qwen3.8 2.4T A95B Quark MXFP4License unverified Language	11.1	—	—	—	2 months ago
Equivalent route	Qwen3.6 35B A3B w8a8 llmcompressorOpen weights Language	7.7	—	—	—	3 months ago
Equivalent route	Mixtral 8x7B Instruct v0.1 w4a16 llmcompressorOpen weights Language	8.0	—	—	—	3 months ago
Equivalent route	Qwen3 VL 8B Instruct w4a16 llmcompressorOpen weights Language	8.5	—	—	—	3 months ago
Equivalent route	Qwen3 VL 8B Instruct w8a8 llmcompressorOpen weights Language	6.9	—	—	—	3 months ago
Equivalent route	Kimi K3 Quark MXFP4 AttnFP8Restricted license Language	10.9	—	—	—	3 months ago
Equivalent route	Llama 3.1 8B Instruct w4a16 llmcompressorRestricted license Language	10.9	—	—	—	3 months ago
Equivalent route	Phi 4 reasoning plus w4a16 llmcompressorOpen weights Language	5.3	—	—	—	3 months ago
Equivalent route	gemma 4 26B A4B it w8a8 llmcompressorRestricted license Language	10.9	—	—	—	3 months ago
Equivalent route	Qwen3 VL 235B A22B Instruct MXFP4Gated access Language	10.9	—	—	—	3 months ago
Represented by Instella MoE 16B A3B Think	Instella MoE 16B A3B SFTRestricted license Language	10.8	—	—	—	3 months ago
Represented by Instella MoE 16B A3B Think	Instella MoE 16B A3B PretrainRestricted license Language	10.8	—	—	—	3 months ago
Represented by Instella MoE 16B A3B Think	Instella MoE 16B A3B DPORestricted license Language	10.8	—	—	—	3 months ago
Represented by Instella MoE 16B A3B Think	Instella MoE 16B A3B BaseRestricted license Language	10.8	—	—	—	3 months ago
Equivalent route	Qwen3.5 397B A17B MXFP4 AttnFP8 V2Open weights Language	12.3	—	—	—	3 months ago
Equivalent route	MiniMax M3 EAGLE3.1 QuarkOpen weights Language	4.9	—	—	—	3 months ago
Equivalent route	Phi 4 reasoning plus w8a8 llmcompressorOpen weights Language	3.5	—	—	—	3 months ago
Equivalent route	gemma 4 12B it w8a8 llmcompressorRestricted license Language	10.8	—	—	—	3 months ago
Equivalent route	tiny qwen3 moe w8a8 int8Open weights Language	7.7	—	—	—	3 months ago
Equivalent route	DeepSeek V4 Pro NVFP4Open weights Language	6.1	—	—	—	4 months ago
Represented by DeepSeek V4 Pro NVFP4	DeepSeek V4 Pro MXFP4Open weights Language	5.4	—	—	—	4 months ago
Equivalent route	Qwen1.5 MoE A2.7B Chat w int8 a int8 symRestricted license Language	10.5	—	—	—	4 months ago
Equivalent route	Llama 3.3 70B Instruct w4a16 llmcompressor asymRestricted license Language	10.4	—	—	—	4 months ago
Equivalent route	Qwen3.5 397B A17B MoE MXFP4Open weights Language	7.0	—	—	—	4 months ago
Equivalent route	MiniMax M3 MXFP4 AttnFP8Restricted license Language	10.4	—	—	—	4 months ago
Equivalent route	MiniMax M3 MXFP4Restricted license Language	10.4	—	—	—	4 months ago
Equivalent route	Qwen3.5 9B da8w8 torchao v0.17.0Open weights Language	2.7	—	—	—	5 months ago
Equivalent route	Qwen3 VL 8B Instruct w4a16 asym torchao v0.17.0Open weights Language	3.7	—	—	—	5 months ago
Equivalent route	Qwen3.5 9B w4a16 tao symgroup torchao v0.17.0Open weights Language	2.8	—	—	—	5 months ago
Equivalent route	Phi 4 reasoning plus w4a16 asym torchao v0.17.0Open weights Language	3.1	—	—	—	5 months ago
Equivalent route	Qwen3.5 9B w4a16 asym torchao v0.17.0Open weights Language	4.3	—	—	—	5 months ago
Equivalent route	Llama 3.1 8B Instruct w4a16 asym torchao v0.17.0Restricted license Language	10.2	—	—	—	5 months ago
Represented by Qwen3.5 397B A17B MXFP4	Qwen3.5 397B A17B NVFP4Open weights Language	5.9	—	—	—	5 months ago
Equivalent route	gpt oss 20b BF16 w8a8 llmcompressorOpen weights Language	5.5	—	—	—	5 months ago
Equivalent route	DeepSeek V4 Flash MXFP4Open weights Language	10.2	—	—	—	6 months ago
Represented by MiniMax M2.7 MXFP4	MiniMax M2.7 BF16Restricted license Language	9.9	—	—	—	6 months ago
Equivalent route	gemma 4 31B it MXFP4Open weights Language	7.7	—	—	—	6 months ago
Equivalent route	MiniMax M2.7 MXFP4Restricted license Language	9.8	—	—	—	6 months ago
Equivalent route	Qwen3.5 35B A3B MXFP4Open weights Language	6.8	—	—	—	7 months ago
Equivalent route	Qwen3.5 397B A17B MXFP4Open weights Language	11.1	—	—	—	8 months ago
Equivalent route	MiniMax M2.5 MXFP4Restricted license Language	9.4	—	—	—	8 months ago
Equivalent route	Kimi K2.5 W4A8Restricted license Language	9.3	—	—	—	9 months ago
Equivalent route	MiniMax M2.1 MXFP4Restricted license Language	9.2	—	—	—	9 months ago
Equivalent route	Qwen3 Coder 480B A35B Instruct MXFP4Open weights Language	3.7	—	—	—	9 months ago
Equivalent route	Qwen3 235B A22B Instruct 2507 MXFP4Open weights Language	8.6	—	—	—	9 months ago
Represented by Instella 3B instruct	Instella 3B SFTRestricted license Language	8.0	—	—	—	last year
Represented by Instella 3B instruct	Instella 3BRestricted license Language	8.0	—	—	—	last year
Represented by AMD OLMo 1B	AMD OLMo 1B SFT DPOOpen weights Language	2.3	—	—	—	2 years ago
Represented by AMD OLMo 1B	AMD OLMo 1B SFTOpen weights Language	6.1	—	—	—	2 years ago

Superseded versions

Board status	Model	Gravity	Context	Input tokens	Output tokens	Released
Superseded by HyLo Llama 6MLA22GDN 64K SFT	HyLo Llama 6MLA22GDN 8K SFTOpen weights Language	4.2	—	—	—	last month
Superseded by HyLo Llama 14MLA14GDN 64K SFT	HyLo Llama 14MLA14GDN 8K SFTOpen weights Language	4.2	—	—	—	last month
Superseded by HyLo Llama 4MLA12GDN 64K SFT	HyLo Llama 4MLA12GDN 8K SFTOpen weights Language	4.4	—	—	—	last month
Superseded by HyLo Llama 8MLA8GDN 64K SFT	HyLo Llama 8MLA8GDN 8K SFTOpen weights Language	4.3	—	—	—	last month
Superseded by PARD Llama 4 1B	PARD Llama 3.2 1BOpen weights Language	7.9	—	—	—	last year

Amd models: common questions

How many AI models has Amd released?

ModelCap tracks 24 canonical Amd models as of 6 November 2026: 24 language models (4 with a public ModelCap Index position). Superseded versions and equivalent serving routes stay listed on this page, marked as such.

What is the best Amd model?

HyLo Llama 14MLA14GDN 64K SFT is Amd's highest-ranked model as of 6 November 2026, at ModelCap Index position #343 with a score of 6.9, followed by HyLo Llama 6MLA22GDN 64K SFT (#344). The table lists every ranked Amd model in Index order.

Are Amd models open-weight?

19 of 24 tracked Amd models publish downloadable weights (open weights). ModelCap labels weight availability from repository metadata, not whether the training stack is open source; the licence class is shown beside each row.

What is the newest Amd model?

HyLo Llama 6MLA22GDN 64K SFT, first listed on 11 September 2026 and currently ModelCap Index #344. New Amd models appear here as soon as they pass automatic identity checks; the launches page tracks releases across every publisher.